SCIENTIFIC REALISM IN THE PHILOSOPHY OF SCIENCE

AND INTERNATIONAL RELATIONS

CANDIDATE: Georgios Evangelopoulos

THESIS SUBMITTED FOR THE DEGREE OF

Doctor of Philosophy

AT THE

London School of Economics and Political Science

Department of International Relations

January 2013

2

SIGNED DECLARATION

I confirm that all the work presented in this thesis is my own.

……………………………………………………………….

Georgios Evangelopoulos

3

ABSTRACT

This thesis sets out to challenge the assumption widely held among IR scholars

that Scientific Realism (SR) is the definite and final interpretation of realism.

The introduction of SR into IR as the latter’s proper meta-theory has been the

incentive for very intense debates about both meta-theoretical and theoretical

IR issues.

I argue that IR has uncritically adopted the strongest version of SR. This can

be seen by comparing the different versions of SR and their anti-realist

alternatives - as these have developed in the Philosophy of Science literature -

to the version of SR which was introduced into IR. It is Critical Realism (CR),

however, a version of SR that originated with Roy Bhaskar, which has

dominated the SR debate in IR. This development has had negative

consequences with respect to the quality of the argumentation about realism in

IR. This notwithstanding, a positive implication of this situation is that IR

scholars who belong in various traditions of thought have criticized SR from

different theoretical angles and thus shed light on many of its shortcomings.

I elaborate on the comments that have been made on meta-theoretical as well

as theoretical issues and come up with my own conclusions about SR and CR.

In this framework, I also deal with two special issues which have arisen from

this debate’s problematique: the question about whether reasons can be causes,

which lies in the foundations of Wendt’s ‘constitutive explanation’, and the

challenge of ‘meta-theoretical hypochondria’, according to which the extensive

concern with meta-theory takes place at the expense of theorizing real-world

political problems. Last, I show, by a way of a novel contribution, that Wendt’s

4

latest undertaking, of a ‘quantum social science’, although compatible with SR,

suffers inconsistencies and misunderstandings in terms of its methodology,

metaphysics, use of quantum mechanics, and application to IR.

This thesis is an interdisciplinary study, which draws upon the Philosophy of

Science, IR and Physics (namely Quantum Mechanics), in order to scrutinize

the use of SR and CR into IR along with its implications for both IR meta-

theory and IR theory.

5

CONTENTS

Declaration 2

Abstract 3

Contents 5

Acknowledgments 8

List of Acronyms 11

1. INTRODUCTION 13

1.1 The Problem 13

1.2 The Argument of the Thesis 15

1.3 Conclusion 32

2. SCIENTIFIC REALISM IN THE PHILOSOPHY OF SCIENCE 34

2.1 Introduction 34

2.2 What is Scientific Realism? 36

2.2.1 The Metaphysics of SR 42

2.2.2 The Semantics of SR 44

2.2.3 The Epistemology of SR 50

2.3 For and Against Scientific Realism: The Arguments 54

2.3.1 Arguments for SR 65

2.3.2 Arguments against SR 76

2.4 Weakening Strategies and Alternatives to SR 97 2.5 Criticism and Conclusion 102

3. SCIENTIFIC REALISM AS DEVELOPED BY ROY BHASKAR 118

3.1 Introduction 118

6

3.2 Two Dimensions of Scientific Practice 121

3.3 Two Rival Philosophies of Science 124

3.4 Transcendental Argumentation: From Philosophy of Science to Ontology 128

3.5 The Intransitivity of Objects of Sense-Perception 131

3.6 The Intransitivity of the World’s Structure 135

3.7 From Natural Causation to Generative Mechanisms 139

3.8 From Generative Mechanisms to the Natural Tendencies of Things: The Domain of the Real 145

3.9 Conclusion 149

4. SCIENTIFIC REALISM IN INTERNATIONAL RELATIONS 152

4.1 Introduction 152

4.2 Some Theoretical Caveats 155

4.3 The SR Debate in IR 164

4.3.1 Wendt on SR 164

4.3.2 Wight on SR 183

4.3.3 Critique: Opponents of SR 202

4.4 Two Important Issues Raised by the SR Debate in IR 233

4.4.1 Causes vs. Reasons 234

4.4.2 ‘Meta-Theoretical Hypochondria’: True or an exaggeration? 238

4.5 Conclusion 246

5. ALEXANDER WENDT’S QUANTUM SOCIAL SCIENCE AND INTERNATIONAL RELATIONS 253

5.1 Introduction 253

5.2 Methodological Issues 257

5.3 Wendt’s Understanding of Quantum Physics 266

5.3.1 Aspects of Quantum Physics Relevant for Wendt’s Theory 267

5.3.2 A Critique of Wendt’s View of Quantum Theory 274

7

5.4 The Mind-Body Problem in Wendt’s Theory 288

5.5 The Obstacles Towards a New Science of International Relations 292

5.5.1 Penrose’s Theory on the ‘Consciousness Issue’ 296

5.5.2 Quantum Consciousness and the Holographic Model of Wendt’s New IR Theory 306

5.6 Conclusion 321

6. CONCLUSION 323

6.1 The Contributions 323

6.2 Directions for further research 328

6.2.1 The agent-structure problem in a nutshell 328

6.2.2 A new approach to the agent-structure problem 330

7. BIBLIOGRAPHY 335

8

ACKNOWLEDGEMENTS

I am deeply indebted to Dr. Spyros Economides and Professor Christopher Hill

whose supervision, continuous support and constructive feedback and criticism

have been not only vital for the completion of this thesis, but also deeply

inspiring and exemplary in their commitment. Working under their supervision

has been a unique intellectual and human experience, which I have enjoyed

immensely. This thesis, which was completed under extremely difficult

circumstances, would not have seen the light of day were it not for their

emotional support during the last three years, which have been extremely trying

for myself and my family. Of course, despite the above, the usual disclaimer

that I alone assume responsibility for the views expressed in this work applies.

I also wish to extend my deepest gratitude to Dr. Panagiotis Papoulias,

Professor Kimberly Hutchings, Professor John Gray, Professor Nicos Mouzelis,

Dr. Aristidis Arageorgis, Mr. Miltiades Theodosiou, Mr. Athanasios Samartzis

and Ms Eleftheria Papoulia for their generosity and willingness to discuss parts

of my work with me and come up with invaluable comments on them. Their

enthusiasm for philosophy has been inspirational and has helped me overcome

various difficulties in the process. It is no exaggeration to argue that without the

help of my dearest friend, Dr. Panagiotis Papoulias, in every aspect of my work,

and without the emotional support of another great friend and scientist,

Professor Charalambos M. Moutsopoulos, I doubt if I could have made it all the

way through. Professor Eleftherios N. Economou, a physicist par excellence,

has helped me understand better quantum mechanics, which I take on board in

9

the last chapter of this thesis, so my gratitude and deepest appreciation to his

person can never be overstated.

Professor Chris Brown is the IR theorist with whom I feel the most

intellectual affinity, although I have never had any personal contact with him

during my studies at the LSE. Through his lectures and, most importantly,

through his books and writings, he has taught me what International Political

Theory actually means and how important it is to the future of IR as a field of

study. First and foremost, however, I have learnt from Professor Brown that one

must relate research in IR Theory to real problems in international political

practice. Therefore, I would like to express on this occasion my intellectual

debts to him.

Professors Kevin Featherstone, William Wallace, Margot Light, Loucas

Tsoukalis, Elias Mossialos, Konstantinos Arvanitopoulos as well as Dr. Peter

Wilson, Dr. Carsten Holbraad and Dr. Andreas Gofas have always stood by me

by generously offering any help I needed during my doctoral studies, so it is

self-evident that I owe them a lot. My PhD research fellows and now very

promising IR academics, Dr. Marjo Koivisto, Dr. Mireille Thornton, Dr. Jill

Stuart, Dr. Kirsten Ainley, Dr. Annika Bolten, Dr. Stephanie Carvin, Dr.

Antoine Bousquet, Dr. Douglas Bulloch, Dr. Robert Kissack, Dr. Tim Oliver,

and, first and foremost, Dr. Felix Berenskoetter, have all been sources of

inspiration to me and their different but productive approaches to IR and FPA

helped me broaden the horizon of my knowledge of these two fields of inquiry.

I wish to thank the Department of International Relations of the London

School of Economics for the financial support it has endowed me during one

year of my doctoral studies as well as for the ‘Gregory Tzirakian PhD

10

Scholarship’, which I held for three years and helped me to go on with my

studies by surmounting great economic obstacles. My sincere and deepest

gratitude extends to Mr. Gregory Tzirakian, the donor for the above

scholarship.

During all these years I have cherished and deeply appreciated the support of

many close and dear friends. These are Professor Dimitrios Kioupis, Professor

Lina Papadopoulou, Professor Nicandros Bouras, Professor Leonidas Xanthis,

Professor George Tsakos, Professor Gregorios D. Siourounis, Professor Stavros

G. Papastavridis, Dr. Karolos Papoulias (President of the Hellenic Republic),

Dr. Vasilios Alevizakos, Dr. Spyridon Kotsovilis, Dr. Vassilios Bogiatzis, Dr.

Giorgos Galanis, Dr. Georgios Ouroumpeis, Dr. Panagiotis Charitos, Mr.

George Karagiannis, Mr. Antonis Lavrantonis, Mr. Manolis Kanakis, Ms.

Xanthi Karadima, Mr. Petros Raptis, Mr. Vlassios Leonardos, Mr. Dimitrios

Kontis, Mr. John Korfiatis, Mr. Dimitris Pipinis, Mr. Georgios Tsakalos, Mr.

Kostas Christopoulos, Mr. Anastasios Tzitizikos, Mr. Phaedon Kydoniatis, Mr.

Demetrios Raptakis, Mr. Konstantinos Rokas, Mr. Vassilios Gavrilis, Mr.

Alexandros Mountanos, Mr. Petros Ochtaras, and Mr. Orestis Papagiannis.

Finally, I would like to express my deepest thanks to my beloved wife

Evangelia, my daughter Ellie, my mother-in-law Kalliopi as well as to my

invaluable sisters, Sophia and Penelope. I am a feminist, by definition, as one

can easily understand! All these women have stood by my decisions and

supported me with the utmost of their strength and affection. Any endowment

of my gratitude and deepest appreciation to them can never be enough.

It is to the memory of my parents, Leonidas and Ellie, as well as to my wife

Evangelia to whom I dedicate this thesis.

11

LIST OF ACRONYMS

ASIR Agents, Structures and International Relations

CR Critical Realism

EFPA European Foreign Policy Analysis

EU European Union

ES English School

FPA Foreign Policy Analysis

IBE Inference to the Best Explanation

IR International Relations

NGO Non-governmental Organization

NMA No Miracle Argument

NOA Natural Ontological Attitude

PMI Pessimistic Meta-Induction

QCD Quantum Chromodynamics

QCH Quantum Consciousness Hypothesis

QD Quasi-Duhemianism

QM Quantum Mechanics

SQUID Superconducting Quantum Interference Device

SR Scientific Realism

12

SSR Structural Scientific Realism

STIP Social Theory of International Politics

UA Underdetermination Argument

13

1

INTRODUCTION

1.1 The Problem

Alexander Wendt, nowadays one of the most well known theorists of

International Relations (IR), first attracted the attention of the IR community

when he published a paper on the agent-structure problem in IR, where he

introduced Scientific Realism (henceforth SR) as a meta-theory for IR (Wendt

1987). He further elaborated on the ideas firstly presented in the

aforementioned article and twelve years later came up with his Social Theory of

International Politics (henceforth STIP), a grand constructivist theory of IR,

which also relies on SR as its meta-theoretical background (Wendt 1999). Both

in the aforementioned as well as in the subsequent and more recent works of

his, which concern his ‘quantum social science’ project, Wendt has used SR as

the meta-theoretical bedrock of his elaborations.

Colin Wight, a strong believer in the virtue of science, after publishing a

number or papers on the philosophy of science and IR, attempted to provide a

‘science of IR’ in his book, Agents, Structures and International Politics (2006)

(henceforth ASIP) (Wight 2006). Being a critical realist, Wight has endorsed

Critical Realism (henceforth CR), the Bhaskarian version of SR, as the proper

meta-theory for IR. Many IR critical realists, like Kurki, Patomäki, Joseph,

Jessop, et al. have followed his example in accepting CR as the meta-theoretical

background of their substantive theoretical studies in IR.

14

All the above pose the following questions: what are actually SR and CR,

and how did they come to be imported into IR? Therefore, I decided to devote

one chapter of my thesis to highlight the SR debate in the philosophy of science

and a second one to a meticulous philosophical analysis of the Bhaskarian

theory of ontology, namely CR, before I go on to discuss the implications that

the introduction of both SR and CR into IR have for the latter’s meta-theoretical

and theoretical considerations. There are, of course, a few IR scholars like

Kratochwil (2000), Smith (2000), Chernoff (2002, 2007, 2009a, 2009b) and

Jackson (2001, 2008a, 2009, 2011) who, equipped with a solid philosophical

background, have aptly criticized SR and CR. Nevertheless, I have not found in

the relevant bibliography any work in which “Scientific Realism in the

Philosophy of Science” (Chapter 2) and “Scientific Realism as developed by

Roy Bhaskar” (Chapter 3) are examined separately and in detail prior to the

critical study of the SR debate within IR, which I discuss in “Scientific Realism

in International Relations” (Chapter 4). The present thesis bridges this

bibliographical void.

Moreover, given that Wendt’s latest intellectual undertaking concerns the

very ambitious programme of building a ‘quantum social science’ that relies on

SR as its meta-theory (its ontological basis being the wave-particle duality in

Quantum Mechanics in this case), I also devote a separate chapter to the

discussion of that project, titled “Alexander Wendt’s Quantum Social Science

and International Relations” (Chapter 5).

To sum up, this thesis is an inter-disciplinary study, which draws upon

Philosophy of Science, IR and Physics (namely Quantum Mechanics), in order

15

to scrutinize the use of SR and CR in IR, along with its implications for both IR

meta-theory and IR theory.

1.2 The Argument of the Thesis

In Chapter 2, I will examine SR as an autonomous pursuit in the philosophy of

science. This is doubly important to IR: first, whereas SR has repeatedly set the

stage for rich and vigorous discussions, in the meta-theoretical context of IR it

has been treated rather casually; second, it is necessary to show that any appeal

to SR, in order to bolster the scientific credentials of an IR meta-theory, cannot

avoid incurring all the associated costs.

The chapter initially examines the main tenets of SR. For the sake of

exposition, I follow Psillos (1999, 2009) and present IR through the prism of

three theses: metaphysical realism, semantic realism and epistemic realism. I

examine metaphysical realism first, since therein lies the major presupposition

of SR. Then I present SR’s semantics, according to which theoretical terms

should be taken literally and not fictionally or metaphorically. Finally, I

examine the epistemological thesis of SR, which claims that science can and

does deliver theoretical truth, very much in the way it delivers observational

truth.

Having discussed the main tenets of SR, I then show how SR needs all three

theses to defend itself against anti-realism. It should be noted that metaphysical

realism on its own is not enough. First of all, a metaphysical commitment to the

existence of the external world does not necessarily entail semantic or epistemic

realism. This allows for double semantic standards between observational and

16

theoretical statements, while keeping to SR’s ontological commitments. One

way to formulate this alternative position is instrumentalism.

According to instrumentalism, theoretical commitments can be abandoned in

favour of the observable consequences of a theory. Instrumentalism therefore

rejects the semantic thesis of SR. However, while acknowledging the

importance of this position, I argue for the indispensability of scientific theories

and of theoretical terms to science. Following Sellars (1963) and Quine (1960),

I insist that theoretical discourse serves explanatory and systematic purposes.

The appeal to the semantic thesis of SR, then, according to which scientific

terms should be taken literally, is necessary for a proper realist position.

Moreover, the metaphysical thesis of SR on its own does not seem able to

account for historical change in science. A systematic divergence between what

there is in the world and what scientists claim there is, would, if substantiated

historically, undermine SR’s robust sense of objectivity. Such a formulation of

the matter construes the realism/anti-realism debate in epistemic terms. I argue

that the epistemological thesis of SR, according to which scientific theory can

and does provide true knowledge, seems necessary to counter any doubts on

science’s objectivity. Consequently, proper realism may be legitimately taken

to require all three SR theses.

Still, putting the metaphysical and the epistemic theses in the foreground

does not necessitate SR: a realist could agree that scientific practice discovers

real entities, but could still deny the literal truth of the theories which describe

them. This type of entity-realism ties in more with experimental practice and

side-steps the question of theoretical truth. However, I argue that this position

would be difficult for a realist to adopt since it is theories that both provide the

17

realist with knowledge of any properties these entities have and allow for the

interpretation of experimental results. Overall, while I find the focus on

scientific practice to be an important contribution of entity-realism, I argue that

it does not ultimately address the issues of the SR debate.

Having defended my formulation of SR, I then turn to an examination of the

arguments for and against it. First, I discuss the arguments used by realists. The

single most important argument, the ‘No-Miracle Argument’ (NMA), employs

the empirical success of science to justify SR via a defence of the epistemic

thesis. I trace the argument’s origins to the work of Smart (1963), Maxwell

(1962, 1970) and Putnam (1975a, 1978). They all argue that empirical success

does not just happen (‘miraculously’) but occurs because theories are true and

because the unobservable entities they posit exist.

I focus on Putnam’s argument specifically. I construe it as an abductive

philosophical argument, which appeals to ‘inference to the best explanation’

(IBE). I show how it can defend adequately the epistemic thesis of SR against

anti-realist objections, by offering an overarching empirical hypothesis as the

best explanation of the success of science, without the need to appeal to any a

priori distinctive philosophical method. Moreover, Putnam’s NMA is capable

of addressing the objections put forward by sceptics like van Fraassen.

However, I conclude that, in order for NMA to provide a successful defence of

the epistemic thesis and thus support SR, it must ultimately appeal to the

metaphysical thesis.

The arguments against SR are two: the Underdetermination Argument (UA)

and the Pessimistic Meta-Induction (PMI). They challenge SR’s epistemic

credentials. The former holds that theories always have empirically equivalent

18

rivals and doubts the realist’s belief in unobservables, because empirical

evidence alone is deemed incapable of fixing theory-choice. The latter holds

that the historical record shows that, contra SR, science is marked by failures

and current science should not be relied on, since theories which have in the

past been considered to be true, are now taken to be false.

I first present UA in detail and show that before one examines the epistemic

issues raised by UA, a stand should be taken on the metaphysical thesis. This is

where PMI comes into relevance. The most promising defence against this

argument is to engage in historical case studies and attempt to reconcile realism

with the historical record. This can be done by identifying not only stable

components that survived disproved theories, but also ones which are often

instrumental for the demise of the theory that they had been part of. Again,

though, even if such case studies were provided, they could never be

conclusive. Moreover, if one chooses to go down the path of examining all past

and present scientific theories, then one risks contextualizing SR’s theses to

such an extent that one may compromise the metaphysical thesis.

In this connection, I discuss Mäki’s ‘doubly local scientific realism’, and

argue that the conceptual flexibility that Mäki’s view entails makes for a very

watered-down version of realism (Mäki 2005). Although it is too early to judge

this ‘new-generation’ SR, I find that at present this watered-down SR lacks the

appropriate vocabulary to provide the argumentative structure one would like to

see and evaluate. Furthermore, I also find that the outcome of the realism/anti-

realism debate has been that one should no longer go uncritically for an

adoption of the standard global conception of SR. This is of utmost relevance,

since one of the central arguments I make in the thesis is that the manner in

19

which SR has been imported and used in IR meta-theory does not appreciate

adequately the grave theoretical implications SR imposes on any strong, global

realist construal of IR. The standard brand of SR mainly adopted by IR meta-

theory is considered to be a definitive and conclusive version of realism; this is

mistaken, since many issues in SR remain unresolved, whereas different

interpretations of realism abound. Picking the strongest version among them

rather uncritically as the definite and final interpretation of realism and

importing it in IR meta-theory is a risky business.

As a response to the issues which I discuss in this chapter, realism and anti-

realism have both almost fragmented. SR must be refashioned. Realists have to

choose what they are realists about. I go on to present briefly both positions

which aim to weaken realism, as well as their alternatives, but under no

circumstances do I exhaust the issue; such a task would be beyond the scope of

the present work. My aim is merely to indicate that there are a variety of realist

theories available.

In conclusion, I examine the fact that even the context itself of the realist

debate generates controversy, due to the idealized conceptions of ‘science’ it

harbours. The role that an idealized conception of science plays within the

realist/anti-realist debate is brought out in detail. I note how the latter might

dissolve if a more refined conception of science were to be used. The adoption

of an unreal, ‘mythological’ conception of science is uncritical and permeates

much of the work in IR meta-theory – notably Alexander Wendt’s and Colin

Wight’s. Consequently, the aforementioned works inherit all the problems that

go with such an adoption.

20

In Chapter 3, I start by arguing that although the influence that SR exerts on

the philosophy of natural sciences remains uncertain, its impact on IR meta-

theorizing has been noteworthy during the last few years. However, as already

said, the complexity of SR is not fully recognized within the existing body of

work of IR. The latter accepts as SR the particular kind of SR that is developed

by Roy Bhaskar, which is also known as CR (Bhaskar [1975] 2008).

Therefore, before I move on to examine the alleged usefulness of SR to IR

as a meta-theory on the grounds of which one can build ‘legitimate’ IR

Theories – something which I will discuss in Chapter 4 of the thesis – it is

necessary to examine the particular SR developed by Bhaskar, especially in

terms of its ontological assumptions – hence my selective and precise use of

Bhaskar’s extensive writings.

In the second section of the third chapter, I present Bhaskar’s conception of

science as a social product directed towards an independently existing objective

world. This allows me to present the distinction between the transitive and

intransitive dimensions of scientific practice. In particular, I explain in what

sense one may speak of transitive objects of knowledge.

In the third section, Bhaskar’s philosophy of science will be set against two

rival accounts concerning the nature of scientific objects: Humean classical

empiricism and Kantian transcendental idealism. For Hume, experience

exhausts all knowledge of the world possible for us; the world is but a surface,

with no underlying structure. For Kant, on the contrary, there is an underlying

structure, but this structure is imposed upon the world by man’s cognitive

activity; it is not an objective feature of the world itself. Bhaskar claims that

both these rival philosophies of science are committed to a common ontology,

21

which he calls ‘empirical realism’. I will show that the crux of this ontology is

the concept of ‘empirical world’, i.e. of the world that is the object of a possible

experience. For Bhaskar, empirical realism leads inescapably to a rejection of

the intransitive dimension of science.

In the fourth section, I will present Bhaskar’s alternative philosophy of

science, for which he uses the label ‘transcendental realism’. There, I am going

to distinguish three levels of analysis: at the level of epistemology, I will show

that transcendental realism is a theory about the objects of scientific practice; at

the level of methodology, I will show that transcendental realism makes uses of

‘transcendental argumentation’; at the level of ontology, I will show that

transcendental argumentation leads to a particular conception of what the world

must be like for scientific practice to be possible.

Bhaskar makes use of two separate transcendental arguments. The first one

focuses on the conditions of possibility of sense-perception, while the second

examines the conditions of possibility of experimental activity. Therefore, in

the fifth section, I will present a novel codification of Bhaskar’s sense-

perception argument, which has the structure of a reductio ad absurdum. This

transcendental argument leads to a categorical distinction between the level of

the ‘empirical’ and the level of the ‘actual’, which also allows Bhaskar to claim

that ‘events’ are to be categorically distinguished from ‘experiences’.

In the sixth section, I focus on Bhaskar’s most famous argument, which

purports to establish the intransitivity of the world’s structure. Contra Hume,

Bhaskar claims that the world should not be viewed as consisting of events and

constant conjunctions of events. The world is not a surface, but has an

underlying structure. However, contra Kant, this structure is not a mere product

22

of human cognitive activity, but an independently existing feature of the world

in itself. Through his transcendental argument which focuses on the notion of

experimental activity, Bhaskar claims to have shown that causal laws of nature

are categorically distinct from the regular sequences of events that occur by

way of experimentation.

Bhaskar’s second transcendental argument makes a substantive claim about

the nature of causation. Natural causation is to be understood as the operation

of some ‘generative mechanisms’ that produce the experienciable actual

phenomena of the world, on the one hand, but which are to be categorically

distinguished from them, on the other. For Bhaskar, what science seeks to

accomplish is to acquire knowledge of these enduring generative mechanisms.

It is this move from natural causation to generative mechanisms that I examine

in the seventh section of the chapter in focus.

In the eighth section, I consider Bhaskar’s next move from generative

mechanisms to the ‘natural tendencies of things’. This examination will allow

me to focus my attention on the specific nature of Bhaskar’s ontology, which is

thoroughly, though not explicitly, Aristotelian in character. As I will claim,

Bhaskar’s philosophy has the paradoxical feature of presenting itself as a novel

philosophy of science, which makes use of Kantian methodology (i.e.

transcendental argumentation) in order to arrive at a non-explicit, yet

undeniable, version of Aristotelian ontology of substances: a thing (Bhaskar’s

equivalent of Aristotelian substance) is to be understood as a natural agent

endowed with particular tendencies (Bhaskar’s equivalent of Aristotelian

habitus) that empower it to naturally perform specific real activities.

23

Returning to the philosophy of science, I will show that, according to

Bhaskar, to state a causal law is to describe a generative mechanism; but to

describe a generative mechanism is to describe the natural tendencies of a thing.

Thus I will explain what Bhaskar means when he talks about a ‘normic’

conditional, which is how a statement of a natural law is to be properly

understood. The distinction between experiences, events and generative

mechanisms leads to a further ontological distinction of the levels of reality: the

empirical, the actual and the real.

To sum up, in this chapter I will show how Bhaskar begins with an inquiry

in the field of philosophy of science and how he comes up with a full-fledged

ontology, which purports to provide us with a richer and deeper image of the

world that goes further and beyond observable patterns of events. However, this

robust new ontology also leads to a novel understanding of the nature of

science. Thus I will next return to the point of departure of this chapter, for it

will then become understandable why Bhaskar’s meta-theoretical principles can

have a profound impact on the way working scientists understand the subject-

matter of their field. Having examined Bhaskar’s take on SR, which, as I

mentioned before, is the prevailing form of SR found in the discipline of IR, I

will turn my attention to the issue of how SR is used in IR theory and what this

use entails.

In Chapter 4, I will examine how the concept of SR and its associated claims

about theory, knowledge, and reality have been employed in IR. After a short

Introduction to the chapter in focus, I will set up some theoretical caveats for

the discussion of SR in IR, by clarifying which of its aspects I am going to

illuminate more than others and why. Next, I will turn my attention first to

24

Wendt and then to Wight, in order to highlight the two different versions of SR

in IR that have been widely discussed within IR, with CR being the one to have

dominated the IR meta-theoretical discourse during the last years. This will be

the start of my critical examination of the SR debate in IR.

I will focus on Wendt in order to prove that there is a thread of thought

concerning SR which penetrates his paper about agents and structures in IR

(Wendt 1987) as well as his next intellectual endeavours, namely his STIP

(Wendt 1999), his paper on the alleged inevitability of the world-state (Wendt

2003), his paper on the state as a person (Wendt 2004), as well as his first

publications on ‘quantum social science’ (Wendt 2006a, 2010) – the latter I

discuss in the last chapter of the thesis. I will argue that Wendt has imported

into IR the strongest of all the versions of SR that I have discussed in Chapter 2.

Specifically, he has imported the version that relies heavily on the following

three premises: first, there is a world out there which exists independently of

our ability to know it or not; second, mature scientific theories must be taken at

their face value; third, the world consists of both observable and unobservable

entities. Wendt embarks upon SR mainly because he needs the notion of

‘unobservables’, so that he can theorize about structures in his attempt to

describe the relationship between the behaviour of the states as units of the

international system and the international system itself (with all its identities,

powers, constraints, etc.). I will underline that he does not want to use the

notion of ‘structure’ instrumentally, therefore he takes refuge in SR and claims

that structures are entities either observable (e.g. material forces) or

unobservable (e.g. culture, ideas); he thus prioritizes ontology over

epistemology and uses the meta-theoretical framework of SR in order to build

25

his own social theory of international politics upon it. I will further show that

Wendt makes this meta-theoretical choice simultaneously with another choice,

that of the agent-structure explanatory theoretical scheme, in order to reject

both the alleged individualist methodology of neorealism as well as the holistic

methodology of the world-system theory as being unable to explain sufficiently

the production, reproduction and change of the morphology of international

politics. In other words, I will show that Wendt’s theoretical fabric consists of

levels which are intrinsically connected in a way that one cannot change any of

them without affecting the others.

In the second subsection, I will deal with Wight who, being a critical realist,

endorses the Bhaskarain version of SR, namely CR (Wight 2006). As I will

point out, he launches an attack against the ‘epistemic fallacy’ in IR, by

rejecting the assumption that reality corresponds to the knowledge we have

about it. Wight prioritizes ontology over epistemology and is all for

epistemological relativism (‘anything goes’) and judgmental rationalism (that

is, in any given case we can choose between competing theories). A lot of

interesting consequences (e.g. SR is rather at odds with the idea of a

‘correspondence theory of truth’) as well as intriguing questions emerge from

these premises. One such question, for instance, is whether Wight’s

epistemological relativism is simply limited to empirical research, in the sense

that epistemology does not or should not play any role in our ontological and

methodological considerations, a subject matter which, as I will show, is still

contested. Moreover, I will highlight the fact that Wight believes that the meta-

theoretical choices are ultimately political.

26

I will close this second section by presenting and critically discussing the

positions of opponents of SR and/or CR, such as Kratochwil (2000), Suganami

(2002, 2006), Chernoff (2002, 2007, 2009a, 2009b), Lebow (2011), Jackson

(2001, 2008a, 2011) and others. A first general conclusion from this discussion

is that many distinguished IR scholars who got involved in the SR debate in IR

cannot distinguish between CR and SR. A second is that Wendt’s theory is

more worthwhile in its theoretical than in its meta-theoretical substance.

In section 3 of the chapter in focus I will deal with two important issues

which have surfaced during the SR debate in IR.

The first one is the question about whether reasons can be causes, too.

According to Hollis and Smith, reasons cannot be causes, since in Explaining

and Understanding International Relations (Hollis & Smith 1990) they endorse

a strict distinction between explaining and understanding in the study of

international politics. On the contrary, Wendt thinks that we need a ‘via media’

between explaining and understanding and that the ‘constitutive explanation’

that he suggests, and which takes causes to be real, should complement ‘causal

explanation’. If this is the case, then reasons can be causes. I will undertake a

philosophical analysis of this issue but at the end I will reach the conclusion

that the reasons vs. causes debate is still open-ended.

The second issue has to do with what Fred Halliday (2000) used to call

‘meta-theoretical hypochondria’ in his attempt to criticize some IR theorists’

extensive concern with ontological, epistemological and methodological

considerations, at the expense of dealing either with theoretical problems which

arise from actual politics or with empirical research. My conclusion is that

interest in meta-theory does not amount to a waste of time, and IR theorists

27

should be aware of their meta-theoretical choices, since the latter determine, at

least to a certain extent, the content of those IR theories which they underpin.

However, one should never forget that the subject matter of IR consists in the

study of real-world political problems, which are inextricably connected with

the daily agonies and hopes of ordinary people for a better life.

The last section of Chapter 4 summarizes in the form of a list the arguments

of the chapter. The overall list consists of twenty two items but, for the time

being, I will point out only two of them: first, I agree with Chris Brown’s

argument that CR can do the IR discipline service only if it can invigorate an

interest in Marxist Studies in IR (Brown 2007); second, I call, along with

Jackson, for a ‘pluralist science of IR’ (Jackson 2011), in which I would,

nevertheless, also like to include – as Onuf suggests – language turn theories

and speech act theories (Onuf 1989, 2012).

One may legitimately think that, after the adoption of SR by IR and its

implications for the meta-theoretical considerations within the field of IR have

been illustrated and criticized in chapter 4, nothing more could be said on this

issue. However, both adherents to and opponents of Wendt’s attempts to

construct a methodological ‘via media’, which could combine positivist and

interpretivist approaches to the study of international politics, have been

challenged anew by his “Social Theory as Cartesian science: an auto-critique

from a quantum perspective” (Wendt 2006a). The purpose of his new research

programme, which has been inaugurated in the aforementioned paper and is

still in progress, is to frame a new ‘quantum social science’, which I discuss in

the fifth chapter of the thesis.

28

I will show that Wendt intends to base his ‘quantum social science’ project

on his ‘quantum consciousness hypothesis’ (QCH), i.e. on the assumption that

human consciousness can be explained through quantum mechanics. This

assumption, furthermore, seems to be related to panpsychism. He argues that

consciousness is a macroscopic quantum mechanical phenomenon in the sense

that it behaves as both a wave and a particle, in compliance with the wave-

particle duality of quantum mechanics. In this case, QCH aspires to resolve the

so-called ‘mind-body problem’, which is one of the most difficult in the history

of both philosophy of mind and science. Wendt was led to this idea from his

belief that the adoption of the word-view of classical physics by social sciences

limits their capabilities to fully explain, and even possibly forecast, the complex

social and political phenomena taking place on the national and international

levels. The solution to this deadlock is, in his view, the development of a

‘quantum social science’ which should be based on his QCH, which holds that

the behaviour of individuals may be explained with the use of the

aforementioned wave-particle duality of quantum mechanics. The

aforementioned explanation presupposes the connection of two distinct physical

concepts, namely the ‘wave-function collapse’ and the ‘interaction with the

environment’. Another point that arises from this discussion and is worth

mentioning is that Wendt does not use quantum mechanics as a metaphor;

instead, he uses it literally in order to provide a completely new and much more

comprehensive social theory.

Wendt’s interest in the study of ‘collective consciousness’ was apparent for

the first time in his paper, “The state as a person” (Wendt 2004). In my view,

this paper marks a turning point for Wendt’s intellectual inquiries into the

29

notion of collective consciousness, since he takes on board the relevant notions

of supervenience and superorganism. Wendt has thus coined the concept of

QCH in order to further explore and elaborate the notion of the collective

consciousness of the state.

I will show that, in order to achieve this goal, Wendt has more recently

invoked another theoretical tool; that of the ‘holographic hypothesis’, which is

based on the ‘holographic principle’ of Astrophysics (Wendt 2010). The main

point of his argument is that consciousness does not end with us but collective

consciousnesses are also possible. Wendt believes this is feasible, if the

‘holographic hypothesis’ is applied when studying the relationship between

individuals, the state and the international system. Such an application would

consider individuals to be simple points in the international system, with each

of them bearing on its own mind or having ‘deposited’ in its own memory all

the information needed for the recreation of the whole international system.

This transcends the distinction between individuals and societies and puts the

classical level-of-analysis problem of IR Theory (Singer 1961) on another

basis.

At this point, I will argue that Wendt’s ‘quantum social science’ is

compatible with SR, since it is anchored on the ontological duality of ‘wave’

and ‘particle’, and prioritizes this ontology in its methodological and

epistemological considerations, according to the demands of the Bhaskarian

version of SR. Furthermore, the answer to the objection that Wendt’s ‘quantum

social science’ belongs to the positivist methodological tradition – him having

thus abandoned his proclaimed ‘via media’ – is that Jackson’s ‘pluralist science

of IR’ allows for a ‘combination’ of SR as a meta-theory with positivism as its

30

methodology. Then I will devote a great deal of Chapter 5 to the discussion of

the methodological, scientific and metaphysical pitfalls associated with the

aforementioned position.

Regarding the methodological issues of Wendt’s ‘quantum social science’, I

will make clear that the invocation of weak analogical arguments, in the

context of which aspects of human and social life are compared to aspects of

the behaviour of quantum matter, is for many reasons erroneous. It is erroneous,

amongst others, first, because the evaluation of an argument that is based on

analogical methodology must take into account all known relevant similarities

and dissimilarities, which clearly does not happen in the case under

consideration. Second, it is erroneous because a genuine explanation cannot be

exhausted by a mere reduction to the familiar, as we have learnt from the work

on scientific explanation produced by Salmon (1992). Furthermore, I will

argue that Wendt’s reduction of the consciousness problem to that of the mind-

body problem, and his attempt to solve both of them through the use of

quantum science, might make sense only in the framework of a metaphor, and

not when the analogy is taken literally.

In order to come to terms with the scientific pitfalls of quantum physics that

are relevant for Wendt’s ‘quantum project’, I will first present some basic

notions and aspects of quantum mechanics. Then I will go on to criticize twelve

points included in Wendt’s paper on ‘quantum social science’ (Wendt 2006a).

By doing so, I will show that Wendt misconceives many notions of quantum

mechanics and misconstrues several physical phenomena. Finally, I will also

clarify the context of the ‘holographic principle’ in Astrophysics, which has

been coined by Susskind. I will do so in order to prove that there are neither

31

similarities nor analogies between the ‘holographic principle’ and Wendt’s

‘holographic model’.

I am going to deal with the metaphysical weaknesses of Wendt’s project

through the examination of the mind-body problem. While one cannot, of

course, ignore the amount of work going on in neuro-science about the mind-

body issue, and the fact that this is hugely ambitious for them, let alone a social

scientist like Wendt, I will argue that Wendt’s quantum project lacks specific

sound metaphysical underpinnings. First, it exemplifies some conceptual

confusion about its metaphysical assumptions concerning matter. Second, its

metaphysical assumptions concerning the relation of mind to matter remain

promiscuously eclectic. This latter point can be deducted from the bibliography

used by Wendt for the grounding of his argument as to how consciousness is

related and reduced to the mind-body problem.

In the following section, I will examine the importance of the above

shortcomings for Wendt’s project regarding the creation of a new science of IR.

This section is divided into two parts. In the first part, I will present and

critically discuss Penrose’s theory of consciousness as developed in his book,

The Emperor’s New Mind (Penrose 1989), and its sequel, Shadows of the Mind

(Penrose 1994a). This is crucial for the consideration of Wendt’s quantum

project, since he based his QCH on Penrose’s work on consciousness. I will

conclude that Wendt makes two mistakes. On the one hand, his selection of

Penrose’s idiosyncratic consciousness hypothesis is eclectic, since he ignores

the vast relevant bibliography on the subject, while, on the other hand, he

misunderstands and misconstrues the basics of the theory itself.

32

In the same section, I will also discuss the implications of Wendt’s QCH and

his ‘holographic model’ for his new IR Theory. I will argue, among others, that

the latter has not reformulated the level-of-analysis problem in a more

comprehensive and ‘productive’ mode. Furthermore, his new quantum project

does not help us understand any better than the already existing theories of IR

and FPA how collective consciousness and, consequently, collective identity

emerge.

My general conclusion is that, as long as some facets of quantum reality – in

particular that of quantum measurement – remain a mystery, they cannot solve

another mystery; that of consciousness. Therefore, Wendt’s new quantum social

science project, which is assumed to be based on some kind of quantum

collective consciousness, is erroneous.

1.3 Conclusion

Alexander Wendt has introduced the strongest version of SR into IR as its

proper meta-theory in order for him to be able to theorize about structures, the

latter being one of the two poles of the agent-structure explanatory scheme he

uses in his STIP. Contrary to Wendt, Colin Wight defends a rather idiosyncratic

version of SR, the Bhaskarian CR, which is not very highly acclaimed by

philosophers of science – this is clearly implied by a careful look at the SR

debate in the philosophy of science literature. I will show why and how these

two developments have had serious implications for the orientation of both

meta-theoretical and theoretical debates within IR.

A lot of distinguished IR scholars have taken issue with SR and/or CR and,

as I will also show, they have provided us with critiques of differing quality.

33

However, what is important is that some of these critiques have been launched

against meta-theoretical considerations proper while others against the

substantive IR theories which the former underpin.

The aim of this thesis is thus the critical discussion of the introduction and

use of SR and CR into IR.

34

2

SCIENTIFIC REALISM

IN THE PHILOSOPHY OF SCIENCE

2.1 Introduction

This chapter is devoted to a critical exposition of the concept of scientific

realism and its associated set of claims about theory, knowledge, and reality. As

a view about science and scientific practice, scientific realism (SR) should not

be confused with realism in philosophy of language, in philosophy of

mathematics, or in political realism and neorealism. Scientific realism is a

mature, autonomous pursuit in the philosophy of science, drawing mostly from

the natural sciences and their history.

As will be described in more detail in Chapter 4, SR has been appealed to

several times in the more recent meta-theoretical debates of IR. Although in the

meta-theoretical context of IR, the scientific realist doctrines have been treated

rather casually (if not simple-mindedly)1, in truth SR has repeatedly set the

stage for varied, rich and vigorous discussions, with numerous arguments,

counter-arguments, and extensively developed strategies over realism in

science. These have revealed that realism is neither as casual and

straightforward a position as it seems, nor a conception that is easy to dismiss.

Consequently, and mostly in order to defend itself, scientific realism has grown,

from quite modest philosophical beginnings, to be the dominant position in

mainstream Anglo-American analytic philosophy of science. Unavoidably, this

1 This claim will become apparent from the analysis in Chapter 4.

35

sophistication has transformed the vivacious realism debate of the 1980s into an

industry of endless exchanges of apparently scholastic arguments and rebuttals,

where the application of more and more advanced technical tools from logic

and semantics is deemed to be a sine qua non for genuine philosophical

progress.2

As a result, no treatment of the realism issue can be both brief and

exhaustive. Hence, although I shall try to go over the most important issues in

as much detail as possible, still I shall not be able to cover all the areas

currently under discussion. The reader should keep in mind throughout (and

hopefully, confirm for herself by the end of the chapter) that SR is a vastly

complicated issue, lively at times, and often repetitive in the argumentative

forms employed in its defence, though always appealing, at least to the extent

that one enjoys displays of skill in argument more than the drawing of definite

conclusions. The reader coming from IR, especially, should be prepared for a

discussion rather more philosophical in kind than she is used to. It should be

apparent, however, that no matter where one comes from, appealing to SR is

not a choice that should be made lightly: as will be seen, realism remains very

controversial in the philosophy of science and one cannot ‘borrow’ from realist

doctrines in order to bolster the scientific credentials of one’s own discipline

without incurring all the associated costs; and, as the IR scholar especially

should appreciate by the end of this chapter, those costs can be quite high.

My next task will be the clarification of the doctrine of SR. Once this task is

over, I will present the familiar arguments found in the literature, first those

defending SR as a reasonable philosophical position (section 3.1), and then 2 Of course, this phenomenon has not gone unnoticed; cf., for example, Rescher (1987): xi. The authors of a more recent work do not hesitate to talk of ‘realist ennui’ (Callender & Magnus 2004).

36

those questioning the coherence of SR as a philosophical view about science

and its historical development (section 3.2). 3 This will produce some

unavoidable repetitions, since I must come back to the core SR doctrines from

different points of view in my examination of how the adherents of SR adapt

their position to every new round of criticism. In the next-to-last section of this

chapter, I mention briefly the various weakening strategies which less

committed realists have pursued as well as the alternatives to SR currently on

offer. Finally, in section 5, I summarize my findings and I offer my own

conclusions. All along, I attempt to portray SR fairly and in the best possible

light as the complex position its adherents take it to be; however, this should

not mislead the reader: my conclusions will mostly be critical.

2.2 What is Scientific Realism?

SR is, essentially, a doctrine about ontological commitment to scientific posits

and theoretical entities.4 The core idea lies in the conviction that mature

science, and the posits of mature science, are generally true in an ontologically

demanding sense; scientific theories and hypotheses refer to real entities

(whether the middle-sized entities of common sense or unobservable entities),

forces, and relations; theoretical claims should be taken literally, potentially

referring to a mind-independent reality whose details science itself studies.

3 Throughout these sections, I follow Psillos (1999). This is a deliberate strategy, considering that among the SR defenders, I have found Psillos’ attitude to be the most representative and the most coherent; his treatment of the issues seems to me to be the most comprehensive one. Psillos tries to cover all the issues, never hesitates to draw detailed distinctions, pursues the realist stance single-mindedly, and seldom allows anything to pass without comment. However, our reliance on Psillos’ work does not imply endorsement of his views, as will be seen in the course of the chapter. 4 For brief (though interesting and at times comprehensive) overviews, see Blackburn (2002), Devitt (2005), Giere (2005), Glymour (1992), Lipton (2005), Psillos (2000a), or Wylie (1986).

37

Most scientific realists ally themselves closely with philosophical

naturalism. 5 Metaphilosophically, this entails that realists tend to prefer

empirical arguments –no different, at least in principle, from the kinds of

arguments scientists put forward– and broadly naturalistic theories of meaning,

reference and knowledge. Last but not least, most scientific realists take it that

‘abduction’, namely ‘inference to the best explanation’ (henceforth IBE), is a

legitimate and reliable method to identify the best (potential) explanation of

natural phenomena.

The debate over SR developed principally in the past thirty years.

Historically, SR became the dominant philosophy of science as a reaction to

Logical Positivism. The logical positivists construed scientific theories and

hypotheses largely as instruments of calculation, or as convenient systems

through which the scientists summarize the empirical regularities of observable

entities and processes. Apparently, unobservables were deemed to be too

‘metaphysical’ for positivist strictures. Instead of taking them to refer to actual

unobservable entities and processes, the logical positivists preferred to explain

them away either instrumentally or reductively (as ultimately referring only to

observable entities and events). However, this instrumentalist-reductive stance

soon came to be challenged: under the guidance of such figures as Karl

Popper6, Grover Maxwell7 and J. J. C. Smart8 in the 1960s, the philosophy of

science took the so-called ‘realist turn’, adopting science as a guide to truth,

5 E.g. Devitt (2005), Psillos (1999). For discussion, see Rosenberg (1996), section 1, and Wylie (1986). Of course, there are exceptions: Niiniluoto (1999) simply abandons naturalism and offers normative accounts instead; Bhaskar (1979) turns to transcendentalism. A very interesting evaluation of naturalism in the philosophy of science is Worrall (1999). Complementary though constructive criticism can be found in Fuller (1991a). 6 Popper (1965). 7 Maxwell (1962). 8 Smart (1963).

38

both ontologically and epistemologically. Out of the ensuing exchange of ideas,

there developed a complex, multidimensional philosophical debate, with

sophisticated arguments for and against SR. The current criticism of SR stems,

mostly, from sceptically oriented empiricists, constructivists, verificationists,

and other so-called ‘anti-realists’. The central issue today has been over how to

interpret scientific theories in order to understand the progress of science.

However, defining SR is not an easy task. As Chernoff notes,

Over the years, SR has taken vastly different forms, including inferential realism

(abductive inference to the best explanation), fiduciary realism (where credibility is

at the core), bivalence realism (according to which all statements are either true or

false), entity realism (restricting SR to claims about theoretical entities and

excluding claims about theoretical truth), theoretical-causal realism (restricting the

positing of theoretical entia to those having causal status), and referential realism

(according to which all entities in scientific theories have genuine ontic status).

(Chernoff 2002: 191)9

Amid all these varieties of SR, one can nonetheless discern a family of views

concerning truth, explanation, reference and progress, with their own

metaphysical, epistemological and semantic dimensions. Realists disagree

among themselves about which of these views should be given priority over

others (which allows anti-realists themselves to multiply by negating selectively

the different realist theses); all of these views, though, take it for granted that

9 In note 6 of the same paper, Chernoff points to Patomäki & Wight (2000) for “half a dozen forms of nonpolitical realism”, to Putnam (1982) for multiple ones, Haack (1987) for nine, and Devitt (1991): 302-3 for “about a dozen”!

39

realism about truth and reality should be preferred over any anti-realist,

constructivist counterparts.

In this section, I shall try to elucidate carefully the basic doctrines, which go

under the name of SR. Following Psillos (1999), I shall adopt the idea that

“going for realism is going for a philosophical package which includes a

naturalised approach to human knowledge and a belief that the world has an

objective natural-kind structure” (Psillos 1999: xix). The presence of such

loaded terms as the above should already alert the reader that SR is not to be

appraised on the basis of locating argumentative fallacies or inconsistencies on

the level of detail; on the contrary, SR seems to be a cluster of interlocking

philosophical theses, making up an apparently coherent whole which, perhaps,

would be better understood if seen as the expression of a stance, or even of a

‘picture’. The guiding thread seems to be a second-order one; namely that,

when it comes to broader philosophical issues in science, realism is the best

explanation we have for the success and coherence of the scientific enterprise.

SR, as a distinct philosophical approach, consists of three theses:

(a) The metaphysical thesis: The world has a definite and mind-independent

structure.

(b) The semantic thesis: Scientific theories should be taken at face-value. They are

truth-conditioned descriptions of their intended domain, both observable and

unobservable. Hence, they are capable of being true or false. The theoretical terms

featuring in theories have putative factual reference. So, if scientific theories are

true, the unobservable entities they posit populate the world.

(c) The epistemic thesis: Mature and predictively successful scientific theories are

well-confirmed and approximately true of the world. So, the entities posited by

40

them, or, at any rate, entities very similar to those posited, inhabit the world.

(Psillos 2000a: 706-7)

Viewed as a set of commitments, the above theses project onto science the

privileged position of providing access to a mind-independent reality, populated

by the posits of our best scientific theories (or entities similar to them), which

do not depend for their existence and nature on the cognitive activities and

capacities of our minds. Therefore, SR defends a picture of science which at

one and the same time embodies a heavy-going metaphysics (perhaps of natural

kinds), a referential semantics, and a rather simplistic (though not simple-

minded), both axiologically and pragmatically, truth-centred notion of scientific

cognition.

Be that as it may, it ought to be understood that these dimensions of realism

are only separated for the sake of exposition. Admittedly, each thesis seems to

connect independently with the others in many significant ways; the definition

of realism, however, presupposes all three. It would really not do to talk simply

of ‘mind-independent entities’ without emphasizing that science provides us

with the rational means to describe and discover the properties of those entities;

that would be the metaphysical thesis without the all-important epistemic

construal, which allows us to rely on the methods of science. Similarly,

defining SR as the position which takes the theoretical and empirical statements

of a science as ‘approximately true’, could be paraphrased away as merely the

metaphysical thesis via the disquotational properties of the term ‘true’ (“S is

true iff S”). Equally, however, one should not make much of the notion of

‘truth’ itself: neither a theory of reference nor a specific account of truth is

41

presupposed in the definition of SR10, only ‘truth’ in a loose semantic sense

which may be cashed out in many different ways, even deflationary ones11. As

we shall see in section 2.2, the purely semantic notion of ‘truth’ in SR is largely

a settled issue: all the parties accept that theoretical claims may be taken as

meaningful and capable of being true or false, without adopting a specific

account of truth by doing so; the debate, therefore, is not over semantics per se,

but over how ontology and epistemology impinge on the semantics. At bottom,

SR is a metaphysical doctrine, with commitments which constrain the realist’s

understanding of science, scientific statements and scientific practice.

Nonetheless, having adopted SR, one may go on and offer a construal of ‘truth’

in detail; indeed, the adoption of SR seems to be compatible preferably with a

non-epistemic notion of truth, which “implies two things: first, that assertions

have truth-makers; and second, that these truth-makers hinge ultimately upon

what the world is like” (Psillos 1999: xxi). On this basis, a realist can oppose a

non-realist over the definition of truth and the appropriate construal of

reference, since the non-realist will usually reject such non-epistemic notions

and restrict truth to matters of warranted assertibility and ideal justification

instead.12

10 For details, see the excellent accounts of this matter in Devitt (2005): 770-1, Stein (1989): 50-1. Mäki (2005): 236 is also very clear about this matter. 11 For example, see Horwich (1998). 12 Having said all this, one would hardly expect any realist to conflate the semantic and the metaphysical (non-epistemic) notions of ‘truth’; no well-informed and sophisticated scientific realist would find it helpful to ignore (at least) twenty years of literature (mostly by Hilary Putnam and Arthur Fine; see e.g. Leplin 1984b, or Psillos 1999, chapter 10) over this matter, nor would it make sense to repeat such a conflation all over again unless there were some original ideas in the offing. Nonetheless, Bhaskar (1993) & (1994) has done exactly this; with no original ideas, oblivious to all the arguments and the hard-won distinctions, he goes on to offer a ‘solution’ to epistemic issues by simply identifying the epistemic and the metaphysical (literally non-epistemic; he calls it ‘ontological’) conceptions of truth. What he comes up with is ‘alethic truth’ (which could be translated as ‘true truth’): “… a species of ontological truth constituting and following on the truth of, or real reason(s) for, or dialectical ground of, things, as distinct from propositions, possible in virtue of the ontological stratification of the world and attainable in virtue of the dynamic character of science” (Bhaskar 1994: 251). I won’t venture

42

I will now attempt to unpack the above theses one by one.

2.2.1 The Metaphysics of SR

Even though SR is not principally a philosophical doctrine about knowledge or

being, its basic presupposition lies in a metaphysical ontology, which has quite

a lot to say about both knowledge and being. This is why one would be well

advised to understand first SR’s specific brand of metaphysical realism before

going on to assess its associated semantic and epistemological doctrines.

SR’s metaphysics involves the claim of mind-independence. How is one to

understand this? It has absolutely nothing to do with traditional idealism and

phenomenalism, the notion that the world consists of mental entities, be they

ideas or actual and possible sense-data. On the contrary, SR’s metaphysics rests

on a materialist (if not physicalist) construal of the entities posited by scientific

theories. Thus understood, idealist or phenomenalist doctrines are repudiated a

priori.

However, what might be prima facie surprising, is that SR’s metaphysics

gives up on common-sense realism as well —in the sense that, according to SR,

common-sense realism (the belief that everyday objects have an independent

existence) is a priori and fundamentally inadequate (if not misguided, since

science is taken to correct the ‘unreflective’ judgements of common sense) as a

form of philosophical realism about the world which science reveals. By

focusing on the theoretical terms and the theoretical mechanisms of a theory,

SR does not merely entertain the view that the empirical adequacy of a theory

somehow mirrors real entities and processes; rather, SR’s metaphysics includes into any explanation for this curious phenomenon; instead I will only point to the devastating critique (though, of course, not a very original one, since all these matters have been exhaustingly debated before for years) of Groff (2000).

43

claims about unobservable entities as much as about observable ones. As a

result, it imposes further requirements on the interpretation of ‘mind-

independence’ than mere empirical adequacy: common-sense realism is implied

by SR, but not vice versa. Assuming the existence of reality, ‘mind-

independence’ extends to the unobservable entities posited by scientific

theories (i.e., for the most part, these entities do exist and do enjoy the

properties science attributes to them); the kind of reality we have assumed is,

after all, the one science depicts, and, ultimately, it is the way science depicts it.

Therefore, we may not know directly what kind of reality we have assumed,

what exactly it is capable of doing (or even what it actually does), but, for SR,

‘common-sense realism’ does not (and cannot) exhaust our accounts of it. This

is SR’s specific brand of realism.13

Thus, SR assumes not only the existence of an external reality, but also a

definite philosophical kind of reality, one which seems to encapsulate a

robust sense of objectivity: the real –whatever it is– is independent of the

cognitive activities and capacities of our minds. It is in this sense that SR’s

metaphysics shapes our notions of knowledge and being: SR does not lay

claim to a specific theoretical picture of reality; it only commits itself to a

philosophical notion, according to which the kind of reality that exists is

knowable scientifically and it is (approximately) as current and ‘mature’

science describes it (i.e. what we have called the epistemic thesis). What is

relevant to the concerns of the ontologist, therefore, is the scientifically and

theoretically describable. The emphasis is throughout on science: “If our

13 For more details on the connection between common-sense realism and SR, see Clarke and Lyons (2002) and Devitt (1999). An interesting common-sense defence of realism is Kitcher (2001a).

44

best science is not our best guide to our ontological commitments, then

nothing is” (Psillos 1999: 70).

Accordingly, the SR version of ‘mind-independence’ abstracts drastically from

scientists’ practices (techniques, limitations, norms), in order to trace out what

is taken to ground them and make them what they are —namely, what actually

makes them true. The epistemic means and the conceptualisations used to

know reality are free to change and evolve; still, according to SR’s core

metaphysics, through these changing and evolving practices, the mind-

independent world emerges as the philosophically relevant arbiter —even

though it is via scientific theories, and scientific theorising in general, that we

discover and map out this world.

2.2.2 The Semantics of SR

If science is taken to be our ultimate arbiter of truth, it goes without saying that

at least one of the main aims of science is to offer true (or approximately true)

claims about how the world is. So, what kind of reality is science taken to

describe? What kind of structure does the metaphysical thesis allow that there

exist? Here, SR has very little of detail to say, although what it does say intends

to capture all that is metaphysically important. Except from pointing out the

unique, unobservable, natural-kind structure posited by scientific theories, SR

resolutely insists that this structure, however in its details, exists objectively and

independently of humans’ ability to know, verify, recognize that [it does]. Instead

of projecting a structure onto the world, scientific theories, and scientific theorizing

45

in general, discover and map out an already structured and mind-independent world

(Psillos 1999: xx).

This way of developing the metaphysical thesis does not concern itself with the

types of entity that exist (material, mental, etc.)14, but rather with what is

involved in claiming that they exist: “… the world is the way the scientific

theory –literally understood– describes it to be” (Psillos 1999: 70, emphasis

added); “Science aims to give us, in its theories, a literally true story of what

the world is like” (van Fraassen 1980: 8, emphasis in the original). One might

interpret this as an expression of “the idea that science is in the truth business”

(Lipton 2001: 347; see especially Lipton 2005), or perhaps, of a view which

“erects current science into a metaphysic and ties scientific realism too closely

to that metaphysic” (Musgrave 1996: 21). Either way, the metaphysical thesis,

in its essence, dictates a literal treatment of scientific theory. At this point, the

need for a literal interpretation of theoretical claims implies that the

metaphysical thesis has to be supplemented by a semantic thesis: matters of

ontological commitment are seen to be the springboard for an elaboration of

SR’s semantics (the propositions of scientific theories are to be taken at face

value).

However, talk of unobservability makes it painfully obvious that scientific

theorizing is not ontologically innocuous: science assigns to the world both

observable and unobservable features. According to SR, this has repercussions

for our interpretation of theoretical claims in science: semantic realism (Psillos

1999: xx) is put forward as the correct stance toward the claims of scientific

theories. Semantic realism is an ‘ontologically inflationary’ philosophical 14 For a recent elaboration of this point, see Psillos (2005).

46

treatment of theoretical terms: “Understood realistically, the theory admits of a

literal interpretation —an interpretation in which the world is (or, at least, can

be) populated by a host of unobservable entities and processes” (ibid). In other

words, the semantic thesis expresses SR’s prior ontological commitment:

theoretical claims, both about observable and unobservables, are neither

metaphorical nor meaningless, neither instrumental nor limited to the behaviour

of observables — theoretical assertions have truth-values, are descriptive, and

reiterate the realist’s metaphysical commitments, only this time in the context

of how one should interpret words. A true assertion has ‘excess content’, which

cannot be fully captured by any reference to observable entities and

phenomena.

Therefore, according to SR, the issue of what makes a theoretical assertion

true should involve the notion of (putative) factual reference: one should

associate SR with the view that truth is a non-epistemic concept, so that

“assertions have truth-makers… and…these truth-makers hinge ultimately upon

what the world is like” (Psillos 1999: xxi). Crucially, this is why the semantic

thesis is not a simple paraphrase of the metaphysical one. Contrary to

appearances, defining semantic realism by using the terms ‘refer’ and ‘true’ and

‘literal interpretation’, on its own does oblige one to take the claims of

scientific theories the way SR actually intends us to do, i.e. as expressions of

essential, metaphysical commitments on our part about the ontological status of

the world. For instance, taking their cue from the philosophy of language,

where the simple ‘disquotational scheme’ of “‘such-and-such’ is true if such-

and-such” transforms all truth-claims into sentences which do not mention

truth, the non-realists may deflate any metaphysical commitments SR entails by

47

translating them away ‘disquotationally’: i.e., as Devitt (2005: 772) remarks,

the mere usage of the terms ‘refer’ and ‘true’ could “also be seen as exploiting

only the disquotational properties of the terms with no commitment to the

robust correspondence relation between language and the world”. Furthermore,

alluding to the explanationist strategy for SR, he continues: “The realist

argument should be that success is explained by the properties of

unobservables, not by the properties of truth and reference” (ibid).

Indeed, there is a long, so-called ‘anti-realist’ tradition in philosophy, which

denies the logical independence of the assertion that an entity exists from the

issue of what constitutes evidence for the truth of that assertion:

The truth of an assertion is conceptually linked to the possibility of recognizing this

truth… If an assertion cannot be known to be true, or if it cannot be recognised as

true, then it cannot possibly be true. (Psillos 1999: xxi-xxii)

Namely, this tradition takes it that the evidential basis for the truth of an

existence-claim, on the one hand, and the semantic relation of designation (i.e.,

factual reference), on the other, should not be distinguished in the way that SR

suggests. ‘Merely’ asserting the existence of something should not be taken as

enough by itself to commit one to SR’s ‘robust sense of objectivity’. Anti-

realists vehemently reject any notion of a ‘non-epistemic’ conception of truth15:

rather, what is issued as existing involves a suitable set of conceptualisations

and epistemic conditions (e.g., being ideally justified or warrantedly assertible)

on our part. The fulfilment of a suitable epistemic/conceptual condition should

15 The standard candidate for which is the correspondence theory of truth or even a causal theory of reference. See section 3.2.

48

be enough, if one is to commit oneself to the truth of a theoretical claim;

putative factual reference is not needed. Semantically, this translates into the

conflation of the evidence-conditions of an assertion’s meaning and its truth-

conditions: it cannot happen that a theory is epistemically right (it meets the

relevant epistemic condition) and yet false — anti-realism precludes (a priori)

this possibility of divergence by advancing an epistemic conception of truth.

This conception of truth, according to which the truth of an existence-claim

is conceptually linked with the fulfilment of (ideal) epistemic conditions,

usually arises from distrust towards theoretical terms. For example, since the

meaning of theoretical terms is not given directly in experience, one might

think that these terms are semantically suspect. SR’s semantic thesis is designed

to deflect the grounds for such suspicion; ultimately, SR’s epistemic and

semantic concerns are parasitic on metaphysical ones; SR, as a doctrine, is

about the world itself and not merely our account of it. For that reason, the

reader is well-advised to keep in mind that the semantic thesis plays a

supportive role to SR’s metaphysics: SR neither stands nor falls with it.

Semantic realism is a doctrine about truth, but “no doctrine of truth is

constitutive of metaphysical doctrines of scientific realism” (Devitt 2005: 771).

Indeed, this is why SR is not, at least essentially, a doctrine about

unobservables per se: rather, theoretical claims about both observables and

unobservables should obey the same semantical standards. Claims about

unobservables rest on contingent grounds: our (epistemic or pragmatic)

difficulties to observe ‘directly’ certain objects of theoretical science.

Unsurprisingly, therefore, scientific realists insist on the truth –conditions of an

existence-claim, as the appropriate semantic standard for how to interpret it.

49

The observable/unobservable distinction should not have any significant

semantic relevance.16

A brief perusal of the writings of major non-realists in the current literature17

should be enough to make it clear that semantic realism is no longer a matter of

debate: theoretical claims, like “photons exist”, are taken by both sides (realist

and anti-realist) to have a literal interpretation, according to which the claim is

taken at face value, and can be either true or false.18 It is far from clear,

however, what the ‘literal’ interpretation of the claim commits us to,

philosophically: it may well be that one admits that the claim has ‘excess

content’ — i.e. content which cannot be paraphrased away as a claim solely

about observables — but does this have any bearing on whether realism is the

correct attitude to science? There does not seem to be any prima facie

incoherence in granting both the semantic and the metaphysical theses, and yet

still be sceptical or agnostic towards scientific theories. After all, merely

admitting that the theoretical claims of science can be true or false certainly

does not commit one to believe the stronger claim that current theoretical

claims may indeed be true.

At this point, the third aspect of SR, the epistemic thesis, is relevant.

16 Indeed, it is not clear what exactly the observable/ unobservable distinction implies for defining scientific realism: “A coarse-grained sense that can be given to scientific realism is to say that it asserts the reality of unobservable entities: there are genuine facts that involve unobservable entities and their properties. But note the oddity of this way of putting scientific realism. I do not, of course, doubt that there are unobservable entities. But isn’t it odd that the basic realist metaphysical commitment is framed in terms of a notion that is epistemic, or worse, pragmatic?” (Psillos 2005: 395). Psillos goes on to offer his own view of the reasons for which the scientific realism debate has focused so much on the observable/unobservable distinction. Later on, he goes so far as to say that “the issue of (un)observability is really spurious when it comes to the metaphysical commitments of realism” (ibid: 396). 17 For example, those of Arthur Fine, Larry Laudan, and Bas van Fraassen. The term ‘non-realists’ in the text has been chosen instead of ‘anti-realists’ because, in this case, only van Fraassen counts as a genuine anti-realist. See section 4. 18 For further details, see section 3.

50

2.2.3 The Epistemology of SR

The need of SR for a third thesis, the epistemic thesis, should be already apparent

to anyone who harbours doubts about the possibility of science to deliver truth.

We may believe in a ‘mind-independent’ world; we may also admit that

theoretical discourse neither reduces to an instrumental interpretation nor limits

itself to talk about observables; nevertheless, why should we accept the far

stronger claim that science does (or even can) provide the ‘royal road’ to truth

and knowledge of how things actually are?19

At this point, perhaps a realist ought to be satisfied with invoking possible

empirical conditions that would warrant attributing some measure of truth to

theories. However, in that case, it is far from straightforward to see how such a

position would issue in rational or warranted belief in the unobservable entities

posited by science (and the assertions made about them). For that reason, SR

puts forward the epistemic thesis: the thesis expresses the realist’s full

commitment to science as a successful, rationally well-founded enterprise: to a

scientific realist, it makes hardly any sense to believe in an objective, ‘mind-

independent’ world, about which scientists make detailed existential and

descriptive claims, and then go on and doubt the existence of the entities they

posit and the truth of what they say of them. On the contrary, according to this

third dimension of SR, science can and does deliver theoretical truth no less

than it can and does deliver observational truth.

However, this does not mean that SR, as a philosophical position on science,

is necessarily uncritical of its current (or past) form. An argument for the realist

interpretation of scientific theories is not ipso facto an argument for believing in

19 See van Fraassen (1998).

51

current (or any) science. Rather, the epistemic thesis concerns itself with

reasons: do we have reasons to doubt that a mature and empirically successful

science delivers truth? It goes without saying that we cannot simply declare a

priori that science has been, or has to be, successful in truth-tracking; why, this

may happen even by accident. Instead, we should take time to go carefully over

its history and all it currently asserts. The epistemic thesis expresses the

scientific realist’s understanding that, in the end, reasons for doubt do not exist:

[The epistemic thesis] of scientific realism intends to stress that it is reasonable, at

least occasionally, to believe that science has achieved theoretical truth. In other

words, the third realist thesis implies that there is some kind of justification for the

belief that theoretical assertions are true (or nearly true), where this justification

comes primarily from the ampliative-abductive methods employed by scientists.

(Psillos 1999: xxi)

If reasons for doubt do not exist, we may as well commit ourselves to what the

current scientific theories assert; this does not mean, though, that we commit

ourselves to everything that current theories assert. Indeed, that we do not (and

should not) do so may be taken as the natural outcome of our critical and

detailed acquaintance with current and past science: having understood the

theoretical statements of scientific theories as expressing genuine propositions

does not require us to provide a blanket endorsement of the claims of science,

since not even scientists themselves adopt such an undifferentiated attitude.

Indeed,

52

[S]cientists themselves have many epistemic attitudes to their theories. These

attitudes range from outright disbelief in a few theories that are useful for

predictions but known to be false, through agnosticism about exciting speculations

at the frontiers, to a strong commitment to thoroughly tested and well-established

theories. The realist is not less sceptical than the scientist: she is committed only to

the claims of the latter theories. […] In brief, realism is a cautious and critical

generalization of the commitments of well-established current theories. (Devitt

2005: 769)

Thus, one must not lose sight of the general philosophical issues at stake: SR is

not about blind trust to science and scientists’ claims (which would amount to

scientism), but about the conviction that well-confirmed and predictively

successful scientific theories should be treated as actual detailed descriptions of

true facts. Equally, SR is not so much about the epistemic scope of science as a

human, social enterprise, as about the philosophical significance of the products

of scientific theories in terms of referring to facts. Indeed, “even if all the

theories we ever come up with are false, realism isn’t threatened” (Psillos

2000a: 708).

By treating science as capable of providing access to how things really are,

scientific realists limit their attention to scientists’ argumentative strategies for

securing well-confirmed predictions; in this way, their true focus is on the

factual descriptions provided by scientific theories, on the finished and well-

founded products of science, and not on how scientists might actually go about

establishing those products. Hence, the label of ‘epistemic’ in the ‘epistemic

thesis’ should not mislead the reader into thinking that, somehow, in passing

from the metaphysical to the epistemic thesis, scientific realists have gradually

53

turned their attention to the scientific practices themselves. On the contrary, the

focus on scientific practice seems, as it did above for the semantic thesis, to

serve the dictates of the metaphysical doctrine itself: to the extent that SR

necessarily employs a distinct metaphysical picture of a ‘mind-independent’

world to understand scientific theories, scientific practice is, equally

necessarily, perceived as producing these theories by accommodating itself to

the ontological features (constraints) of this world.

Therefore, one might be tempted to call the label of ‘epistemic’ a misnomer;

indeed, Psillos himself comes to the same conclusion when hard-pressed to

explain exactly how much weight should one assign to the metaphysical thesis

when defining SR:

Perhaps it was unfortunate that I called the last dimension of scientific realism

‘epistemic’. I was carried away by sceptical anti-realist attacks on realism. I would

now call it: the factualist thesis. (Psillos 2005: 396)

The point, then, is that SR, even when concerned with actual scientific practice,

restricts itself to how the practice contributes to the tracing out of true facts.

Indeed, one should not forget that SR is designed as a “ready-made

philosophical engine” (Fine 1986a: 177) to interpret mature and successful

science20; inevitably, this ‘engine’ takes in the sophisticated, complex, social

and material traditions of inquiry we call science, and outputs a rather

homogeneous collection of putative factual descriptions, produced by an ‘army’

of single-minded intellects advancing on truth. Admittedly, this is a caricature

20 “If science does succeed in truth-tracking, this is a radically contingent fact about the way the world is and the ways scientific method and theories have managed to ‘latch onto’ it” (Psillos 2000a: 713).

54

of science as much as of SR; nevertheless, it certainly chimes well with another

claim from a prominent scientific realist, a claim which one might equally well

characterize as a caricature of science in these post-Enlightenment days: in

response to the interpretation of SR as the combination of semantic thesis and

an axiological thesis (along the lines of “science aims for true theories”), Psillos

laments that, were we to restrict SR to this characterization, then “all the

excitement of the realist claim that science engages in a cognitive activity that

pushes back the frontiers of ignorance and error is lost” (Psillos 2000a: 708).

2.3 For and Against Scientific Realism: The Arguments

Not unreasonably, philosophical worries about the commitments of SR abound.

As noted above, one is well advised to keep the issues of the ontological, the

semantic, and the epistemic dimensions of SR distinct. Even though one may be

ontologically committed to the existence of an external world, it is optional to

be a semantic, or an epistemic realist. Clearly, belief in a mind-independent

world does not entail that one can have true knowledge of this world; it is

perfectly coherent for a metaphysical realist to insist on adopting a strong

instrumentalist stance when interpreting current scientific theories, even the

‘mature’ and successful ones. This might allow for double semantic standards

(one for observational statements and another for theoretical ones), without

however bringing any obvious damage to a scientific realist’s ontological

commitments. So, what decides the appropriate stance of a realist? Are all

realists forced to adopt SR if they are to honour their ontological commitments?

Instrumentalist claims have developed mainly in response to the above

dilemma. Apparently, for an instrumentalist, ontological commitments have no

55

normative bearing on a realist’s theoretical commitments: it is entirely

reasonable to take the meaning of scientific theories to be fully captured by

what they say about the observable world; theories should be seen as

instruments for the organization, classification and prediction of observable

phenomena. Consequently, serious attempts were made on the part of

instrumentalists to provide further support for this idea.

Historically, this position was taken to locate the terms of the debate in the

interpretation of the theoretical statements versus the observational ones.21

Using mostly arguments from logical analysis and mathematical logic, Craig’s

theorem (Craig 1956) was called upon to argue that theoretical commitments in

science were in principle dispensable; they could be eliminated without loss in

the deductive connections between the observable consequences of the theory.

This was eventually formulated as “the theoretician’s dilemma” by Carl

Hempel22: given Craig’s theorem, the instrumentalist’s reasoning developed

roughly as follows: if theoretical terms were supposed to have no factual

reference, then their truth-conditions would not obtain; without truth-

conditions, a full explication of scientific theories would allow that one

dispenses with theoretical terms entirely. And if this could actually be achieved,

then no further question would arise over the factual interpretation of a

theoretical statement (or over any commitment to irreducible unobservable

entities).

A careful analysis of the reasoning above shows that two assumptions

generate the dilemma if accepted uncritically: on the one hand, a sharp

distinction between the theoretical language and the observational language has 21 In addition, the observable/unobservable distinction took center-stage in the definition of SR. See note 16. 22 Hempel (1958).

56

been assumed without argument; on the other, scientific theories are conceived

as establishing solely a deductive systematisation of observable phenomena.

Both assumptions are highly implausible; this became apparent in the 1960s,

when both the sharp dichotomy and the rather narrow view of theories were

severely challenged.

Following Pierre Duhem23, a holistic view was put forward, according to

which no observational term is devoid of theory (in other words, all observation

is ‘theory-laden’). A scientific theory’s terms function in interconnection with

the rest of the language of science, and so none is anchored to the world of fact

alone. In this sense, all observation terms have what one might call a

‘theoretical’ component; observations no longer represent the ‘ultimate’ data of

science. Unless one conventionally stipulates what is to count as an

‘observation term’, the distinction between ‘observation terms’ and ‘theoretical

terms’ is impossible to maintain. As Maxwell (1962) pointed out (among many

others), in practice the sharp distinction presupposes a uniform use of terms

which simply cannot exist. Hence, strictly speaking, there can be no purely

observation terms, and theories do not represent merely the scaffolding of

knowledge, to be used and then dismantled; rather, theoretical language and

observations form an organic whole of interacting associations.

In addition, no theory has the restricted scope assigned to it by the adherents

of instrumentalism: theories are multi-purpose tools, if tools they are; they

allow for grand generalisations of observable phenomena, based on inductive

arguments and inductive systematisations. They establish connections among

observables, connections which go beyond mere deductive systematisations.

23 Duhem ([1906] 1954).

57

Moreover, theories help us determine what happened by postulating an

explanation of why it happened: they do not provide solely for empirical

adequacy or correct verifiable predictions; they also characterize explanatory

success. Therefore, one cannot simply identify the explanatory power of a

theory with its predictive power. As such, scientific theories and theoretical

terms are indispensable to science. Once these points were realized, there

followed a series of indispensability arguments by Sellars (1963) and Quine

(1960) (among others), which suggested that theoretical terms were

indispensable for any mature and successful system of laws to organize and

explain why observables obey the empirical laws they do.

Consequently, the instrumentalist interpretation of theoretical discourse in its

original form (either as the non-assertoric treatment of theoretical statements, or

as disguised talk ultimately about observables) proved very hard for anti-realist

philosophers to hold on to. Thus, theoretical discourse came gradually to be

taken as irreducible and meaningful by all sides of the debate; giving up the

attempts to paraphrase theoretical statements into observational discourse, and

admitting that theoretical terms are indispensable to a ‘mature’ and successful

science, allowed philosophers to reach a kind of consensus, at least on this

matter.24 As the alert reader will have noticed, however, the instrumentalist

challenges above were mainly to the semantic thesis of SR; they had very little

to say on either the metaphysical or the epistemic thesis. Indeed, this is why

semantic realism is no longer a matter of contention (as noted in section 2.2.2).

Nevertheless, mere assent to a non-instrumentalist interpretation of scientific

theories does not warrant or compel the title of realism. Mere opposition to

24 For an excellent account of how this came about, see Psillos (1999), chapters 1-3.

58

instrumentalist semantics does not entail commitment to realist metaphysics,

either. Hence, once again I have come full circle back to the unclarity which I

began with: in what sense is one’s realism scientific realism? The extent of a

scientific realist’s philosophical commitments is not yet sufficiently clear.

Granted, one can take SR’s thrust to lie in fully subscribing to the

metaphysical thesis; its other dimensions may be construed as further

developments of it, or even supportive of SR’s ontological kernel. The major

philosophical worry, then, is whether SR, by committing itself to such a strong

thesis from the beginning, has rendered itself immune to the empirical claim

that, historically, science has not been devoid of misses in its record of truth-

tracking; indeed, this worry has been the source of current anti-realist criticism.

It goes without saying that scientific realists allow for the possibility of a

divergence between what there is in the world and what scientists claim there

is. As noted in section 2.2.1, scientific realism does not entail uncritical

acceptance of current science. It is felt, however, that weakening our

commitment to the accounts of current science compromises the robust sense of

objectivity scientific realists are after. This robust sense of objectivity is taken

to be necessary if a number of anti-realist or social constructivist views about

science are to be blocked. Therefore, the terms of the debate have focused on

our stance toward the accounts provided by science: if one grants that a mature

and predictively successful scientific theory is assertoric and ontologically

inflated, why should one not take it as generating true (or approximately true)

beliefs and knowledge about the world? Is it not irrational to keep an agnostic

or sceptical stance toward theories in that case?

59

Apparently, putting the matter in this way construes the realism/ anti-realism

debate in epistemic terms; namely, this time any anti-realist challenge has to be

to the epistemic thesis of SR. Psillos puts it thus:

[I]s there any strong reason to believe that science cannot achieve theoretical truth?

That is, is there any reason to believe that, after we have understood the theoretical

statements of scientific theories as expressing genuine propositions, we can never

be warranted in claiming that they are true (or at least, more likely to be true than

false), where truth is understood realist-style? (Psillos 2000a: 713-4)

Inevitably, once we assent to both metaphysical and semantic realism, it does

indeed seem hard to abstain from adopting the scientific realist’s

epistemological rationale: it seems entirely reasonable to grant that the

scientific enterprise (construed as pursuit of truth) should give rise (at least

sometimes) to genuine knowledge of the world. SR’s epistemic realism entails,

therefore, ‘epistemic optimism’, according to which, science not only aims at

truth and does the best it can in order to achieve this aim, but also that this aim

is in principle achievable —the ampliative-abductive methods of science can

produce theoretical truths about the world and deliver theoretical knowledge. In

this sense, SR is ‘presumptuous’ (Psillos 2000a: 707): the modest claim, that

“there is an independent and largely unobservable-by-means-of-the-senses

world, which science tries to map” (ibid) should not be enough for SR; what

one should aim for is

the more presumptuous claim…that, although this world is independent of human

cognitive activity, science can nonetheless succeed in arriving at a more or less

60

faithful representation of it, enabling us to know the truth (or at least some truth)

about it. (ibid)25

It is significant to make clear that this in no way assigns to philosophers of

science the task of deciding which scientific theories should be heeded and

which should not, any more than it allows them to distinguish the absolutely

true theories from the absolutely false ones in current science:

[I]t seems obvious that what realism should imply by its presumptuousness is not

the implausible thought that we philosophers should decide which scientific

assertions we should accept. We should leave that to our best science. Rather,

presumptuousness implies that theoretical truth is achievable (and knowable) no

less than is observational truth. (Psillos 2000a: 714)

Indeed, Psillos, instead of pressing home the ontological aspects of SR, opts for

the epistemically strong claim, that “[T]he ampliative-abductive methods of

science are reliable and can confer justification on theoretical assertions” (ibid),

since nothing less can establish the sought-after connection between the mere

possibility that certain epistemic conditions “warrant attributing some measure

of truth to theories —not merely to their observable consequences, but to

theories themselves” (Leplin 1997a: 102), and the guarantee that this possibility

may be actualised in fact. Anything else than that, simply

fail[s] to guarantee that this possible connection may be actual (a condition required

for the belief in the truth of a theoretical assertion), [whereas] any attempt to give

25 Psillos notes that the terminology of ‘modesty’ and ‘presumptuousness’ is borrowed from Wright (1992): 1-2.

61

such a guarantee would have to engage the reliability of the method which connects

some empirical condition with the truth of a theoretical assertion. Hence, the

defence of the rationality and reliability of these methods cannot be eschewed.

(Psillos 2000a: 714)

Again, this makes it quite plausible that the debate over SR should be

formulated in epistemic terms. These stronger claims epistemologically

distinguish SR from both contemporary versions of empiricist philosophies of

science (which deny the possibility of rationally justifying belief and/or

knowledge about unobservables) 26 and neo-Kantian, constructivist views

(which deny the possibility of epistemic access to a mind-independent world)27.

Consequently, it is no wonder that many realists have chosen to go for the

stronger version as above.28

Nonetheless, one should not forget that the epistemic dimension of SR is just

that: one of its dimensions, not the entire position.29 SR does not consist in one

single doctrine; therefore, the reader is advised to keep in mind SR’s status as a

family of interconnected theses, and hence, that the metaphysical and the

semantic theses are always (and crucially) lurking in the background, even

though the debate over SR takes place in epistemic terms. That this has further

epistemological significance can be seen in the philosophical diversity of SR

itself. As will be noted in section 2.3, realists take some interest in actual

scientific practice, although they usually adopt a rather a priori realist attitude

26 For example, van Fraassen (1980); see especially his (1998). 27 Explicit examples here are more difficult to find, though Kuhn (1970) might be understood as one such; see Boyd (1983), section 4, Devitt (1984) or Hoyningen-Huene (1993). 28 For example, Boyd (1983), Devitt (2005), Leplin (1997a), Psillos (1999), chapter 4, (2000a), Trout (1998), chapter 3. It is unfortunate that the otherwise learned and deep paper of Chernoff (2002) misses exactly this point (Chernoff 2002: 193) when he claims that Leplin (1997)’s ‘modest’ version of SR is representative of realism. 29 This point is strongly made in Devitt (2005): 770-1.

62

in ontologically interpreting it. However, conjoining the metaphysical thesis

with the epistemic thesis while downplaying the semantic one, allows a realist

to accept that scientific practice discovers real entities and processes

(unobservable or not), yet deny the literal truth of the theories in which

descriptions of these entities and processes are embedded. Indeed, doubting that

any current theory provides correct descriptions and thus sidestepping the

question of theoretical truth, has been amply defended by Hacking (1983) and

Cartwright (1983); it might even be that such a position, at least prima facie,

with its emphasis on entities which may transcend theories (and theory-change)

comes up more realist than standard SR. Furthermore, this kind of realism ties

up more clearly with actual experimental practice, where manipulation of

entities seems more decisive than any corroboration of scientific theory in the

abstract. Indeed, thanks to such positions, there has been a renewed interest in

the philosophy of experiment, the philosophy of models which scientists use in

their endeavours, and a gradual shift toward scientific practices as the primary

focus of the philosophy of science.30

For this reason, such positions have drawn quite a lot of attention from both

realists and anti-realists. Much discussed has been Hacking’s conception of

realism, which distinguishes between entity realism and theory realism

(Hacking 1983: 27). The one is about theory-transcendent entities, and the other

about the truth-valuedness of scientific theories, respectively. Hacking claims

that it is no great departure from the spirit of scientific realism to deny theory

realism while asserting entity realism. This refocuses attention from realist

interpretations of theoretical discourse to realist conceptions of scientific

30 See, for example, Franklin (1986), Galison (1987), Gooding (1990), Gooding et al (1989). For the turn to actual scientific models, see Giere (1998) and Morgan & Morrison (1999).

63

practice, specifically of the laboratory practice. Experimental practice would be

inexplicable, it is claimed, not if certain theories were not accepted as true, but

if specific unobservable entities did not really exist; experimenters are taken to

be dealing with laboratory phenomena caused by real entities, and this provides

them with good reasons to be realists about theory-transcendent entities instead

of theories. In any case, it may be (and it usually is the case) that experimenters

are dealing with entities common to several theories and models, so it is not

always clear which theory (if any) should be credited with the honorific of

‘true’:

[O]ne can believe in some entities without believing in any particular theory in

which they are embedded. One can even hold that no general deep theory about the

entities could possibly be true, for there is no such truth. (Hacking 1983: 29)

Even people in a team, who work in different parts of the same large experiment,

may hold different and mutually incompatible accounts of electrons. That is

because different parts of the experiment will take different uses of electrons.

Models good for calculations on one aspect of electrons will be poor for others.

Occasionally a team actually has to select a member with a quite different

theoretical perspective simply in order to get someone who can solve those

experimental problems...There are a lot of theories, models, approximations,

pictures, formalisms, methods and so forth involving electrons, but there is no

reason to suppose that the intersection of these is a theory at all. (Hacking 1983:

264–5)

We are completely convinced of the reality of [e.g.] electrons when we regularly set

out to build —and often enough succeed in building— new kinds of device that use

64

various well-understood causal properties of electrons to interfere in other more

hypothetical parts of nature… We design apparatus relying on a modest number of

home truths about electrons, in order to produce some other phenomenon that we

wish to investigate. (Hacking 1983: 266)

After all, theoretical views undergo change; whereas, posited entities (e.g.

electrons), if real, survive theory-change. In addition, experimenters do things

with these entities: they do not just detect and measure them; they also

manipulate them. Hence, “engineering not theorizing is the best proof of

scientific realism” (ibid: 263).

Is this a coherent position for a realist to adopt? It seems quite difficult to put

theories aside when, at the same time, it is only theory that can provide the

realist with knowledge of any properties those entities may have.31 On the other

hand, these properties do come with the theoretical presuppositions of the

experimenters; why should one identify them independently with theories? Are

the approaches of all groups involved in a modern experiment tuned, as it were,

to each other, so as to form a theory?32

Perhaps the truly radical aspect of entity realism is the attention it devotes to

experimental practice instead of abstract scientific theory. Indeed, a lot of non-

trivial work seems to be indispensable for successful experimental processes

and results.33 Standard SR, however, usually involves an implicit ‘filtering out’

of the details of scientific practice, since the experimenters’ methods belong to

a context perhaps too ‘raw’ and unrefined for rational consideration. For

31 See Clarke (2001), Gelfert (2003), Harré (1996), Morrison (1990), Nola (2002), Resnik (1994). 32 See Franklin (1986), Galison (1987). 33 See Collins (1985), Gooding (1990), Gooding, Pinch & Schaffer (1989) and the references in the previous note.

65

example, according to SR, genuine understanding of putative facts about

electrons ultimately requires the elaboration of theories, which extend far

beyond what is being used and needed in laboratory life.

Nonetheless, it should be noted that even if experimenters need some

theoretical description of the entities they manipulate, this does not entail that

they use comprehensive theories about them. In the final analysis, what the

relative independence of experimentation from theory should make us accept is

only that experimental manipulability bestows the right to causal talk. In its

turn, talk about causes sets essential constraints upon further theory

construction. And it is the final product of this process —theory— that realism

should concern itself with and reason about. Entity realism, therefore, cannot

be fully divorced from theory realism —or so standard SR claims. Once again, I

have come full circle to the idea that standard SR requires the full panoply of

the metaphysical, the semantic and the epistemic theses.

2.3.1 Arguments for SR

Primarily with the defence of the epistemic thesis in mind, current realists have

focused on justifying their realism by appealing (not unreasonably) to the

empirical success of science. Consequently, several so-called ‘explanationist’

strategies for SR have been developed and refined over the last few decades,

based on arguments from the empirical success of science. Amongst these

arguments one of the most highly regarded is the IBE of the empirical success

of science, the so-called ‘No-Miracle Argument’. 34 This argument has

34 Boyd characterizes it as “the argument that reconstructs the reason why most scientific realists are realists” (Boyd 1983: 49).

66

undergone quite a lot of refinements35, been roundly criticised by Laudan

(1981) and others (see section 3.2), and yet is constantly reformulated and

reaffirmed. It is usually traced back to three variants, as follows:

(i) Smart’s36 ‘No Cosmic Coincidence Argument’,

Is it not odd that the phenomena of the world should be such as to make a purely

instrumental theory true? On the other hand, if we interpret a theory in the realist

way, then we have no need for such a cosmic coincidence: it is not surprising that

galvanometers and cloud chambers behave in the sort of way they do, for if there

are really electrons, etc., this is just what we should expect. (Smart 1963: 39)

(ii) Maxwell’s37 argument from the empirical virtues of realistically interpreted

theories,

As our theoretical knowledge increases in scope and power, the competitors of

realism become more and more convoluted and ad hoc and explain less than

realism. For one thing, they do not explain why the theories they maintain are mere

cognitively meaningless instruments are so successful, how it is that they can make

such powerful, successful predictions. Realism explains this very simply by

pointing out that the predictions are consequences of the true (or close true)

propositions that comprise the theories. (Maxwell 1970: 12)

35 See, for example, Putnam (1978), Boyd (1984), (1985), (1990), Hacking (1983), Devitt (1991), Ladyman (2002), chapter 7, Leplin (1997a), Psillos (1999). 36 Smart (1963). 37 Maxwell (1970).

67

and (iii) Putnam (-Boyd)’s38 ‘No Miracle Argument’ (henceforth NMA), which

is arguably the most famous attempt at a purely realist explanation of the

success of science:

The positive argument for realism is that it is the only philosophy of science that

does not make the success of science a miracle. That terms in mature scientific

theories typically refer (this formulation is due to Richard Boyd), that the theories

accepted in a mature science are typically approximately true, that the same terms

can refer to the same even when they occur in different theories – these statements

are viewed not as necessary truths but as parts of the only scientific explanation of

the success of science, and hence as part of any adequate description of science and

its relations to its objects. (Putnam 1975a: 73)

Note that the three arguments do not address the same camp of anti-realists:

Smart and Maxwell are concerned to rebut semantic instrumentalists, hence

they defend the semantic thesis, whereas Putnam argues for both the semantic

and the epistemic theses; he wants to defend both the claim that theoretical

statements may genuinely refer (and so they can neither be merely instruments,

nor be reduced to non-theoretical facts), and that the theories themselves are

approximately true (a logically stronger statement). Note also how Smart talks

of cosmic coincidence and Putnam of miracles. On this point, both agree that

SR does not allow for either cosmic-scale coincidence or miracles: observable

phenomena do not just happen to be, and do not just happen to be related to one

another in the way suggested by the theory. It is because theories are true and

because the unobservable entities they posit exist that the phenomena are, and

38 Putnam (1975a), (1978).

68

are related to one another, the way they are. This agreement should not mislead

one, however, into thinking that the argumentative structures they use are the

same. As Psillos notes,

[...] Smart’s ‘no cosmic coincidence argument’ relies on primarily intuitive

judgements as to what is plausible and what requires explanation. It claims that it is

intuitively more plausible to accept realism over instrumentalism because realism

leaves less things unexplained and coincidental than does instrumentalism. Its

argumentative force, if any, is that anyone with an open mind and good sense could

and would find the conclusion of the argument intuitively plausible, persuasive and

rational to accept —though not logically compelling. (Psillos 1999: 73)

Putnam’s argument, by contrast, is not a ‘general’ philosophical argument, and

certainly not a priori. Specifically, it is a philosophical argument of the

abductive kind, i.e. it appeals to IBE. As such, it assumes that abduction is a

reliable method of inference. However, the argument makes this assumption on

rather naturalistic grounds: it takes it that IBE is the scientists’ own method of

producing approximately true theories, and since, typically, these theories have

been arrived at by means of IBE, IBE is reliable.39 For that reason, the scientific

realist feels entitled to appeal to abduction and specifically IBE when invoking

realism as the best explanation for the empirical success of science. I shall

discuss in section 3.2 how this appeal to IBE is central in the defence of realism

–especially when put forward in a naturalistic spirit.

Maxwell, on the other hand, differs from both Putnam and Smart: Maxwell

focuses on the epistemic virtues which realistically interpreted theories are 39 Indeed, Putnam (1975a) and Boyd (1973) had already argued that IBE is how scientists form and justify their beliefs in unobservable entities.

69

taken to enjoy, virtues such as simplicity, lack of ad hocness, and explanatory

power, and then submits that statements lacking such virtues are less likely to

be true than those that have them. Furthermore, as Psillos explains in some

detail40, Maxwell’s argument involves an appeal to Bayesian probabilities,

specifically to the ‘prior probability’ (reflecting the initial plausibility ranking)

of competing hypotheses for science’s success. Arguing on the basis of

epistemic virtues of theories, and assuming that realism is better supplied with

epistemic virtues than anti-realism, Maxwell infers that scientific realism has

much higher prior probability than instrumentalism. Thus, he attempts to offer

grounds for Smart’s “primarily intuitive judgements as to what is plausible and

what requires explanation” (Psillos 1999: 71), reaching the same conclusion as

he: one should choose realism over instrumentalism. Roughly, then, whereas

Smart’s argument relies on seemingly a priori plausibility judgements (which

may turn out to be distinctively philosophical), Putnam’s proceeds rather

naturalistically (capitalizing on the abductive-ampliative methods scientists

already use). In that sense, “Maxwell’s argument is the ‘bridge’ between

Smart’s a priori argument and the subsequent Putnam-Boyd naturalistic

version” (Psillos 1999: 74).

However, taking into account that semantic realism seems to be contested no

longer, plus the fact that current, sophisticated forms of anti-realism dispute

exactly Maxwell’s assumption (of realism as being better supplied with

epistemic virtues than anti-realist contestants)41, it is prudent to focus on

Putnam’s No Miracle Argument for SR. Indeed, Smart’s and Maxwell’s

40 Psillos (1999): 74-5. 41 As Psillos takes pains to make clear, the modern, sophisticated anti-realist (van Fraassen, for instance) adopts a position which “…starts at precisely where Smart’s and Maxwell’s arguments stop”, Psillos (1999): 77.

70

arguments, as originally formulated, seem to beg what is currently the main

question: they assume that once semantic realism has been established, the

epistemic thesis (that we have reasons to believe that our scientific theories are

successful in truth-tracking) can be granted without further argument –almost

as if belief in the truth of our theories were inevitable and rationally

compelling. Furthermore, they do not assign adequate weight to novel

predictions42: one should put extra stress on the fact that “only on a realist

understanding, novel predictions about phenomena come as no surprise”

(Psillos 2000a: 715). In other words,

Since false theories can issue correct predictions, the realist needs some way to

locate instances where he can be sure that truth is responsible for success. Since the

prediction of built-in results allows the possibility that a theory is false but still

successful, it seems the realist would be better off arguing that the real miracle is

success where the result played no part in a theory’s development; where the result

is independent of the theory. (Rich 1999: 512)

Fortunately, NMA seems designed to defend exactly what is currently the main

focus of debate, namely the epistemic thesis of SR, and it can do so in a non-

question-begging manner: NMA can contribute to the defence of realism as an

overarching empirical hypothesis supported by the fact that it offers the best

explanation of the success of science, while at the same time rejecting any

appeal to an a priori, distinctive philosophical method. Finally, NMA can be

reformulated in such a way as to incorporate the highly theory-dependent

dimension of scientific and experimental enquiries, and thus can be called on to

42 Leplin (1997a) is the latest realist book-length attempt to make sense of novelty.

71

address the worries of current selective sceptics (such as van Fraassen). In this

version, though still an abductive argument, NMA has a different explanandum:

scientific methodology itself. Perhaps the best reformulation along these lines is

Boyd’s:

According to the realist, the only scientifically plausible explanation for the

reliability of a scientific methodology that is so theory-dependent is a thoroughly

realistic explanation: scientific methodology, dictated by currently accepted

theories, is reliable at producing further knowledge precisely because, and to the

extent that, currently accepted theories are relevantly approximately true. (Boyd

1990: 223)

This kind of argument also seems perfectly in line with the epistemic optimism

of realism, so let us examine briefly how exactly NMA supports the epistemic

thesis. The main issue for the realist is to provide rational grounds to believe

that scientific methodology leads not only to correct predictions and

experimental success, but also to (approximate) truth and access to the real.

Most scientific realists43 adopt an IBE-based reading of NMA, generalising

from the realist understanding of successful instances of IBE-explanatory

reasoning in science, to a grand abductive argument for the reliability of IBE

itself as the way to secure truth. The common pattern throughout is that to

believe in the reliability of abductive reasoning is reasonable, since it tends to

generate approximately true theories. Therefore, to address the defence of the

epistemic thesis one proceeds in three steps:

43 Boyd (1981), (1983), Devitt (2005), and Psillos (1999), (2000a). For complementary accounts see Putnam (1978): 18-9, Leplin (1984): 203, 205, 217, and Musgrave (1988): 232-4, 239.

72

(I) Firstly, they adopt an empirical-naturalistic approach: the realist takes it

that the problem is ultimately empirical, not a distinctively philosophical one;

accordingly, an empirical hypothesis seems to be in order. Realism is put

forward in exactly this manner: “[The] defence of realism cannot be a piece of

a priori epistemology, but must rather be part and parcel of an empirical-

naturalistic programme which claims that realism is the best empirical

hypothesis of the success of science” (Psillos 2000a: 717).44 Thus, treating the

empirical success of scientific theories as a ‘radically contingent’ and

‘experimental’ fact, it is suggested that realism accounts in the best way for that

fact: “According to the distinctly realist account of scientific knowledge, the

reliability of scientific method as a guide to (approximate) truth is to be

explained only on the assumption that the theoretical tradition which defines

our actual methodological principles reflects an approximately true account of

the natural world” (Boyd 1983a: 71).

(II) Second, having assumed the (no longer disputed) semantic thesis, they

identify how scientific methodology achieves approximate truth and

instrumental reliability: realists suggest that, typically, reliable theories are

arrived at via abductive reasoning, so that abduction is not only a legitimate

scientific method but also a reliable one –it produces approximately true

theories. Therefore, the theories that are implicated in the (best) explanation of

the instrumental reliability of first-order scientific methodology should be

accepted as (relevant approximately) true. This is, of course, an empirical

44 Psillos attributes the originality of this move to Putnam (1978): 19.

73

claim: “This conclusion is not meant to state an a priori truth. The reliability of

abductive reasoning is an empirical claim, and if true is contingently so”

(Psillos 1999: 80).

(III) Finally, in the same naturalistic spirit as before, the realist may employ the

same methods used by scientists themselves to provide an argument for the

hypothesis of realism (provisionally taking any apparent circularity as a non-

vicious one): since abductive reasoning tends to generate approximate truth (i.e.

typically, the successful theories have been arrived at by means of IBE), IBE is

legitimated as a cogent manner of reasoning, and hence, instances of such

reasoning should be taken to be reliable, i.e. leading to truth. In that case then,

NMA can go through successfully as an instance of IBE, a sort of ‘meta-

abduction’: one supports, via IBE, the global hypothesis of realism as the best

explanation for the success of science –(I) above– by appealing to local and

particular instances of IBE-explanationist strategies for securing truth –(II)

above. “[S]uccessful instances [of IBE] provide the basis (and the initial

rationale) for this more general abductive argument” (Psillos 1999: 79): If

particular empirically successful theories are approximately true, and have been

arrived at (by scientists) via IBE, then, realist philosophers, using IBE, may

safely conclude that the empirical adequacy and success of science is best

explained by the fact that science achieves truth. Therefore, science can deliver

theoretical truth –and the epistemic thesis can be safely and rationally asserted.

Note how in (III), IBE is used twice: once for successful scientific reasoning,

and a second time for NMA itself. Note also that NMA is not simply a

74

generalisation over scientists’ use of IBE; rather, it serves the interests of a

broader realist epistemology of science:

Although itself an instance of the method that scientists employ, NMA aims at a

broader target: to defend the thesis that Inference to the Best Explanation… is

reliable. […] So, what makes NMA distinctive as an argument for realism is that it

defends the achievability of theoretical truth. (Psillos 1999: 79)

Both of these features have drawn severe criticism from non-realists: the double

appeal to IBE has been charged to be viciously circular, since by already

employing IBE, NMA presupposes what is to be shown, i.e. that IBE is a

reliable inferential method (“…the objection does alas appear to show that the

No Miracle argument preaches only to the converted: it has no probative force

for those who are not already inclined to use inference to the best explanation”,

Lipton 2001: 349). Furthermore, why should one take it that standard scientific

reasoning and explanatory power have any evidential weight in the practice of

philosophy itself?45

In defence, scientific realists appeal to naturalism and try to deflect both

charges by admitting the fallibility of the argument on both counts: indeed, IBE

could turn out to be unreliable, but realists have no reason to consider it

unreliable, and, in any case, this does not concern NMA but IBE independently

of the former, whereas the debate is over NMA (Psillos 1999: 85-6); NMA is

no worse than attempts to defend modus ponens and inductive rules (ibid: 89); 45 For standard formulations of both objections, see Fine (1986a), (1986b), (1991), Laudan (1981), van Fraassen (1980). Lipton (2001) and Douven (2001) give an excellent evaluation of the circularity objection, both pro and con. The interesting aspect of Lipton’s evaluation is that the NMA reasoning “is the drawing of a general moral from the prior commitment to the truth of specific theories”, (351, emphases ours). The implications of this remark are examined in the final section of this chapter.

75

and thirdly, NMA is not an a priori argument, it works in a non-foundational

and naturalistic context, where judgments are informed by highly theory-

dependent “backgrounds” and no a priori justification is intended (Boyd 2000).

Finally, and most importantly, although NMA is explicitly based on the

particular instances of explanatory reasoning in science to defend the more

general claim that science provides access to truth, this does not allow for

unrestricted application of the argument to empirical success. On the contrary,

application of the argument to predictive success simpliciter is widely — and,

in my view, rightly— regarded as unsound.46

An objection recently put forward47 and “even harder to shake” (Lipton

2005: 1267), is that the NMA itself is fallacious: it ignores “adverse base rates”

(Howson 2000), a statistical fallacy to which we are often quite prone. This is a

sophisticated and complex objection:

The miracle argument trades on the intuition that most false theories would have

been unsuccessful. (Here one is considering the set of possible theories, not just

those that have actually been formulated.) The intuition is correct, but the inference

to truth is fallacious. Of course if all false theories were unsuccessful, then all

successful theories would be true. However, from the fact that most false theories

are unsuccessful, it just does not follow that most successful theories are true. One

way to see this is to start not with falsehood, but with success. Given the constraint

of success, we know that a true theory is one possibility; but given the

underdetermination of theory by data, we know that there are also many false

theories which would have enjoyed that same success. Most false theories would

not meet the constraint, but many would, so alas it looks as though most successful

46 Cf. Musgrave (1988): 231. 47 See Callender & Magnus (2004), Lipton (2005).

76

theories are false! Success may be a good test in the sense that it has a low false

positive rate, since most false theories are unsuccessful. However, that is not

enough to show that most successful theories are true. ... What seems to be going on

is that the miracle argument encourages us to assess the reliability of empirical

success as a test for truth by estimating its false-positive rate (the chance that a false

theory is successful), a rate we rightly judge to be very low. However, we ignore

how incredibly unlikely it would be that, prior to testing, a given theory should be

true. This has the effect of hiding from our view all those other theories that would

be just as successful even though they are false. (Lipton 2005: 1267-8)

In a sense, then, NMA re-expresses the realist’s global commitments to science

as a road to truth, only now supported, in a biased way, via local

considerations. Once again, therefore, it should not be forgotten that realist

commitments rest above all on the metaphysical thesis —neither NMA nor the

epistemic thesis stand or fall simply on their own; they are part of a

philosophical ‘package’ which has to be compared in toto against competing

empiricist or constructivist packages.48

2.3.2 Arguments against SR

There are two major challenges to scientific realism: the Underdetermination

Argument and the so-called Pessimistic Meta-Induction. Both arguments

challenge SR’s epistemic credentials. The first begins from the claim that theories

always have empirically equivalent rivals, i.e. empirical evidence is never enough

on its own to determine the ‘one true theory’; the second begins from an historical

record of theories ultimately rejected, despite apparently enjoying at one time just

48 Cf. Devitt (2005): 774.

77

the same kind of success which, according to the realist, indicates truth in current

theoretical counterparts.

Although this pair of arguments has a venerable history (going back at least to

Duhem and Poincaré), it would be better understood if placed in the context of the

current debate over the epistemic thesis of SR: they both aim at securing

epistemic implications, namely that belief in theory is never warranted by the

evidence, and even if it were, the historical record of science shows that

theoretical, scientific knowledge is neither cumulative nor progressive.

The Underdetermination Argument against realism is motivated by a strong

scepticism toward belief in unobservables. It should be noted that it has no one

clear source49, although responding to it has been of continuous interest to

realism’s defenders50. The argument springs from the observation that more than

one theory can accommodate the very same body of empirical evidence; the

conclusion drawn is a doctrine of underdetermination of theory by all possible

evidence: empirical evidence alone is deemed incapable of fixing theory-choice.

A very simple and powerful formulation belongs again to Boyd:

Call two theories empirically equivalent just in case exactly the same conclusions

about observable phenomena can be deduced from each. Let T be any theory which

posits unobservable phenomena. There will always be infinitely many theories

which are empirically equivalent to T but which are such that each differs from T,

49 Here it is worth mentioning Duhem (1906), (1908), and Quine (1960), (1975), among others. The argument in its modern form belongs to the Princeton philosopher Bas van Fraassen in his famous 1980 book, The Scientific Image. Van Fraassen calls his view Constructive Empiricism (see section 4). Cf. van Fraassen (1976) or (1980), chapter 3. Current debates include Douven (2000), Kukla (1994a), Laudan & Leplin (1991), Leplin (1997b), (2000), Okasha (1997).

50 Bergström (1984), Boyd (1973), Devitt (2002), Earman (1993), Glymour (1980), Hoefer and Rosenberg (1994), Kukla (1994a), (1996b), (1998), chapter 5 & 6, (2000), Ladyman (2002), chapter 6, Laudan (1990), Laudan and Leplin (1991), Leplin (2001: 397), Newton-Smith (1978), (2000a), Psillos (1999), chapter 8, Stanford (2001), Wilson (1980).

78

and from all the rest, in what it says about unobservable phenomena (for formalized

theories, this is an elementary theorem of mathematical logic). Evidence in favour

of T's conception of unobservable phenomena ("theoretical entities") would have to

rule out the conceptions represented by each of those other theories. But, since T is

empirically equivalent to each of them, they all make exactly the same predictions

about the results of observations or experiments. So, no evidence could favour one

of them over the others. Thus, at best, we could have evidence in favour of what all

these theories have in common –their consequences about “observables”– we could

confirm that they are all empirically adequate —but we could not have any

evidence favouring T's conception of unobservable theoretical entities. Since T was

any theory about unobservables, knowledge of unobservable phenomena is

impossible; choice between competing but empirically equivalent conceptions of

theoretical entities is underdetermined by all possible observational evidence.

(Boyd 2002)

The core idea of the argument, then, lies in the straightforward attribution of

indefinitely (possibly infinitely) observational equivalent rival theories to a

successful theory. The upshot is the lack of sufficient rational grounds for

preferring any theory to its rivals.

However, as Boyd himself notes, this ignores the fact that the theory and its

postulated rivals could still yield different observational predictions when

supplemented by appropriate auxiliary hypotheses (about the initial conditions,

the instruments in the laboratory, the mathematical formalism, etc.), in which case

there could be observational evidence favouring one over the other. Indeed, this is

part of ordinary scientific practice: scientists routinely supplement theories with

well established auxiliary hypotheses in order to obtain observational predictions

79

from them. Moreover, as Laudan and Leplin (1991) argue, these auxiliary

assumptions vary over time: they are both defeasible and augmentable. Therefore,

what counts as the set of observational consequences of a theory also varies over

time. So it is not entirely clear what one should take for the extent of the

argument’s application: what should count as ‘observations’ and what as

‘theoretical statements’? This is why Boyd concludes that

… it is probably best to think of the underdetermination argument as applying, not

to "small" theories, but to “total sciences”, large-scale conceptions of the world that

might represent the total scientific conception of the world at a time. Such a

conception would already contain all of the auxiliary hypotheses which were

legitimate by its lights, so the problem just mentioned does not arise. In this revised

form the underdetermination argument says that — whatever our best scientific

conception of the world may be at any given time — we will never have any

evidence that it embodies knowledge of unobservables. (ibid)

If that were the case, then the key word in all this would be ‘unobservables’: the

underdetermination argument would seem to rely on a strict dichotomy between

observables and unobservables, since empirical equivalence is taken to entail

different theories with the same observational consequences. However, it has been

argued that such a distinction is more of a matter of degree rather than an

absolute, ontological dichotomy.51 Moreover, it may well be that in the future the

different theories will be reduced to (or replaced by) a third, more general theory

which is empirically viable but not underdetermined (has no conceivable rivals);

or at some point, there are successful observational predictions which allow one 51 Cf. Maxwell (1962), the analysis of this paper in Creath (1985), and Kukla (1998), chapters 8 & 9.

80

to reject the rival theory. So, actually proving underdetermination may require

hard theoretical work in the context of actual scientific practice. However,

it is probably very difficult to provide actual examples of genuine

underdetermination. If this can be done at all, it is surely a task for scientists rather

than philosophers. But it seems unreasonable to expect scientists to concern

themselves with such a task. Scientists are expected to contribute to the growth of

science. They are not supposed to provide empirically equivalent alternatives to it.

Hence, the thesis of underdetermination will perhaps never be more than a plausible

conjecture. (Bergström 1984: 354)

Both objections certainly make sense as part of a defence of realism, especially if

one takes into account the metaphysical dimension of SR: it has already been

stressed repeatedly in the previous sections, that, according to SR, the ontological

status of an entity should not be determined by its epistemic role in our theoretical

accounts —observational status included. The metaphysical thesis should be

accepted or rejected before one examines the epistemic issues raised by the

underdetermination argument. With the metaphysical thesis in the background,

the observational/theoretical distinction appears to be too anthropocentric and too

shifting (theory- and technology-dependent) to provide cogent grounds for

rejecting the realists’ attitude: first of all, one might counter that the special

epistemic role of the senses, on which the argument’s emphasis on observation is

based, derives from the fact that they are the only detectors we have built in to our

bodies; furthermore, what counts as ‘unobservable’ can be revised (and refined)

through the use of instruments and procedures whose justification is theory

dependent.

81

This last point is crucial for the defence of realism if one focuses on extra-

experimental standards for theory assessment, such as explanatory virtues or

grounds for inductions to new experimental cases. It is certainly not the case that

every theory-choice has to rely on experiment —else there would be a ‘crucial’

experiment every time a scientist makes a non-empirical judgment. Therefore, if

underdetermination covers both rival theoretical claims and auxiliary hypotheses,

then scientists have to appeal to extra-experimental standards to adjudicate

between competing empirically equivalent theories, e.g. the role of explanation as

an evidential standard.52 Thus, by rehabilitating the notion of what counts as

evidence for a theory (so as to go beyond mere observational predictions)

knowledge of ‘unobservable phenomena’ is possible.53 The problem, then, with

the underdetermination argument lies with its excessively limited and unrealistic

construal of ‘scientific evidence’ in actual scientific practice: we want theories

that predict, but also explain, in a balanced and homogeneous manner, both

observable and unobservable phenomena.

Summing up: the above considerations do not amount to a straightforward

refutation of the underdetermination argument, but they certainly point to the

proper context for its assessment; namely, the broader realist understanding of

theory-choice both in current practice and in past history. If one has already

adopted an instrumentalist, anti-realistic construal of theories, the

underdetermination argument seems an attractive and cogent objection to the

realist position. Closer attention to actual scientific practice, however, shows that

knowledge about unobservables may sometimes be obtained, on the grounds that

52 For more details, see the excellent account in Boyd (2002). 53 See Boyd (1983), Lipton (1993), McMullin (1984). Another evidential standard may be the indirect manipulation of ‘unobservables’ (Morrison 1990, Nola 2002); indeed, I have already mentioned (in section 3) entity-realism (Hacking 1983) as one such line of thought.

82

‘evidence’ covers more than simple predictive success; it also provides

explanatory power, which takes one beyond observations and instrumental

successes.

Nonetheless, this alone is not enough to establish the credentials of such a

strong position as SR; the above remarks show that, at least in this case, realism

may tally better with ‘official’ scientific practice (where, occasionally, one seems

to be able to choose, from a pool of empirically equivalent theories, the one

theory which is more epistemically warranted than the others), but they certainly

do not make realism reasonable across the spectrum. Indeed, being able to choose

among empirically equivalent theories would have no bearing on the SR position,

if it came about that one have no reasons to take seriously SR’s epistemic

commitments. Inevitably, this brings me back to the debate over the epistemic

thesis, specifically over the manner according to which one should understand

science’s aims and development.

At this point, the main obstacle to the realist position is the historical record of

science itself: the available historical evidence can be taken to show that scientific

theories have been repeatedly overthrown as false, despite enjoying explanatory

and predictive success (at the time); therefore, one might (meta-)inductively argue

that belief in the realistic interpretation of current or even future theories is

unwarranted. This inductive reasoning (from past referential and truth-tracking

failures of science, to lack of truth and referential success of current theories), is

the so-called Pessimistic Meta-Induction (henceforth PMI). The argument

generally contrasts old theories, which even though they were accepted, and the

evidence for them was quite persuasive, nevertheless, mostly turned out to be

incorrect in the unobservables they posited. From this, the anti-realist concludes

83

that with regard to the theories we currently accept, we should believe that

(probably) most of them are likewise incorrect in the unobservable entities they

posit.

PMI’s main proponent has been Larry Laudan, who has repeatedly pointed out

that

[b]ecause [most past theories] have been based on what we now believe to be

fundamentally mistaken theoretical models and structures, the realist cannot possibly

hope to explain the empirical success such theories enjoyed in terms of the truth-

likeness of their constituent theoretical claims. (Laudan 1984: 91-2)

Indeed, Laudan’s own 1981 article “A Confutation of Convergent Realism” has

become the locus classicus for the Pessimistic Meta-Induction (Laudan 1981).

Comparing a variety of past theories to current ones, its central point aims to

discredit the realist belief that current theories converge (even approximately)

to how the world really is. Consequently, the realist’s explanation of science’s

empirical success (for example, the No Miracle Argument for SR) cannot stand

up in the face of historical facts.

Note that in order to motivate the argument, Laudan takes for granted (at

least provisionally) the common ground of both realists and anti-realists,

namely the point of view of our current theories as a standard of comparison for

understanding and evaluating past theories. However, the argument trades on

the realist view of our current theories in order to conclude that the realist view

84

is untenable. Perhaps there is some tension here.54 For this reason, Laudan's

PMI should be roughly reconstructed as a reductio (Lewis 2001, Psillos 1996):

1. Assume that success of a theory is a reliable test for its truth.

2. So most current successful scientific theories are true.

3. Then most past scientific theories are false, since they differ from current

successful theories in significant ways.

4. Many of these past theories were also successful.

5. So successfulness of a theory is not a reliable test for its truth.

To substantiate this claim, Laudan cites the following long list of theories as

evidence: the crystalline spheres of ancient and medieval astronomy, the

humoral theory of medicine, the effluvial theory of static electricity, the

‘catastrophist geology’, the phlogiston theory of combustion, the caloric theory

of heat, the vital forces theories of physiology, the electromagnetic ether, the

optical ether, the theory of circular inertia, and the theories of spontaneous

generation (Laudan 1981: 33). All of these are currently taken to be fictions.

From this list of examples (which, Laudan says, “could be continued ad

nauseam”55), he concludes that a theory may be empirically successful without

being approximately true. Therefore, the realist’s focus on empirical success is

radically weakened: how can truth-likeness explain success if even empirically

successful theories turned out to be false?

54 See Leplin (1997): 141-5. 55 Laudan (1981): 224 in Balashov & Rosenberg (2002).

85

It goes without saying that realists have roundly criticized this argument on

many grounds.56 The range of the defence is quite astonishing, varying from a

straightforward denial of the empirical successfulness of Laudan’s examples57,

to the elaboration of quite intricate causal theories of reference58. In between,

one may find a sophisticated eclecticism in what may be realistically taken as

true in the examples of Laudan.59 This diversity is not surprising, considering

that the pessimistic meta-induction is said to be a “hard blow” to the

explanationist defence of realism (Psillos 1999: 101), “the most powerful

argument against scientific realism” (Devitt 2005: 784), and “among the niftiest

arguments for scientific anti-realism” (Lange 2002: 281).

The best defence against the argument has focused on reconciling the

historical record with realism. Engagement in historical case studies60 usually

involves showing that the abandoned theoretical components are not essential

for the explanatory and predictive success of past theories: the realists should

be selective in what they are realists about, both in terms of theories (only

‘mature’ theories should count) and of theoretical constituents (only those that

have been retained in our current theories should be taken into account). If the

theoretical components that survive theory-change are those that are responsible

for the abandoned theories’ successes, then the realist should attempt to identify

56 Boyd (1983), Carrier (1991), (1993), Cummiskey (1992), Devitt (1984), Hardin & Rosenberg (1982), Kitcher (1993), chapter 5, Ladyman (2002): 236-52, Lange (2002), Leplin (2004), McMullin (1984), (1987), Psillos (1996), (1999), chapter 5, Worrall (1994b). 57 McMullin (1987); Worrall (1994b). 58 Cummiskey (1992); Kitcher (1993); Leplin (1979). Causal theories of reference trace their origins to Kripke (1972) and Putnam (1975c, 1983b). See also Devitt (1981). This turn to causal theories of reference does not preclude older, so-called ‘descriptivist’ theories of reference; on the contrary, Psillos himself opts for such a one (1999: chapter 12). See also Niiniluoto (1999). 59 Kitcher (1993): 140-9; Psillos (1999), chapters 5 and 6; Worrall (1989). 60 For interesting realistic reconstructions, or reconstructions which argue for SR, see, for example, De Regt (2005), Kitcher (1993), Achistein (2002), Psillos (1994), (1999), chapter 6, Saatsi (2005), Trout (1994), Worrall (1994b). For corresponding criticism, see Chang (2003), Stanford (2003b).

86

those theoretical constituents of abandoned scientific theories, separate them

from others that were more like speculative extras, or ‘idle posits’, and did not

play an explanatory role in predictions, and, finally, demonstrate that those

components which made essential contributions to the theory’s empirical

success were retained through theory-change.61 Psillos, for one, acknowledges

the argument’s force while trying to learn from it and adapt his position

accordingly:

Laudan, realists should say, has taught us something important: on pain of being at

odds with the historical record, the empirical success of a theory cannot issue an

unqualified warrant for the truth-likeness of everything that the theory says. Insofar

as older realists have taken this view, they have been shown to be, to say the least,

unrealistic. Yet, it would be equally implausible to claim that, despite its genuine

success, everything that the theory says is wrong. The right assertion seems to be

that the genuine empirical success of a theory does make it reasonable to believe

that the theory has truth-like constituent theoretical claims. … [I] suggest that the

best way to defend realism is to use the generation of stable and invariant elements

in our evolving scientific image to support the view that these elements represent

our best bet for what theoretical mechanisms and laws there are. (Psillos 1999: 109)

Devitt, too, admits that

Scientific Realism already concedes something to the meta-induction in exhibiting

some scepticism about the claims of science. It holds that science is more or less

61 This has been the line of argument developed in detail in daCosta and French (2003), Kitcher (1993), Ladyman (2002), and Psillos (1999). See Bishop (2003), Bishop & Stich (1998), Douven & Van Brakel (1995), McLeish (2005) and especially Stanford (2000), (2003a) and (2003b) for incisive criticism.

87

right but not totally so. It is committed only to well-established theories not exciting

speculations. It leaves room for a theoretical posit to be dismissed as inessential to

the theory. … In the light of history, some scepticism about the claims of science is

clearly appropriate. The argument is over how much, the mild scepticism of the

realism, or the sweeping scepticism of the meta-induction. (Devitt 2005: 784)

However, both Devitt and Psillos find that, if realism is to accommodate

Laudan’s historical examples, the features of the realist position should change:

How should realists circumscribe the truth-like constituents of past genuinely

successful theories? I must first emphasize that we should really focus on the

specific successes of certain theories… (Psillos 1999: 109)

Settling the argument requires close attention to the historical details. (Devitt 2005:

785) … [This] may be manageable. For, the anti-realist must argue that the

historical record shows not only that past failures are extensive but also that we

have not improved our capacity to describe the unobservable world sufficiently to

justify confidence that the accounts given by our current well-established theories

are to a large extent right. This is a hard case to make. (ibid: 788)

Note that both answers presuppose realism in order to make sense. They

amount to prescriptions for a certain sort of historiographical analysis. Does

this mean that the scientific realist should really stop arguing and do ‘realist’

history instead? Should the debate be over realist historiography instead of

realism? This is unclear, but it would certainly be a serious change of focus:

instead of grand claims and grand theories, one is advised to go local (“we

should really focus on the specific successes of certain theories…”), and

88

procceed case-by-case, to uncover those theoretical constituents which “really

fuel the derivation” (Laudan & Leplin 1991: 462). What is needed are careful

case-studies that will:

• Identify the theoretical constituents of past genuine successful theories that

made essential contributions to their successes; and

• Show that these constituents, far from being characteristically false, have

been retained in subsequent theories of the same domain.

If all kinds of claims that are inconsistent with what we now accept were

essential to the derivation of novel predictions and in the well founded

explanations of phenomena, then one cannot possibly appeal to their truth-

likeness in order to explain empirical success. Then, Laudan wins. However, if

it turns out that the theoretical constituents which were essential are those that

have ‘carried over’ to subsequent theories, then the ‘pessimistic [meta-]

induction’ gets blocked. Settling this issue requires detailed study of some past

theories that qualify as genuinely successful. (Psillos 1999: 110-1)

It is important to mention at this point the fact that arguments contra SR have

been repeatedly drawn from the history of science; Kuhn’s (1970) and

Hanson’s (1958) arguments are two of the most notable. Today the original

arguments seem to be considered mostly peripheral to the debate; they seem to

have been assimilated to the objections coming from the camp of sociology of

science and of scientific knowledge (see section 5). In any case, both Kuhn and

Hanson argue from historical examinations of the growth of science to the

conclusion that neither method nor observation supports the realist conception

of scientific growth: theory-dependence affects scientific knowledge to such a

89

degree that nothing is left for realists to hold onto when theoretical perspectives

change (Kuhn’s scientific revolutions). Boyd (2002) calls their challenges the

‘first version’ of the ‘Neo-Kantian Challenge’ to SR (under the ‘second

version’, he places the strategies and the alternatives to SR I shall discuss in

section 4 of this chapter) and offers a cogent critique of this line of argument,

treating their arguments as calling for a different theory of reference of

theoretical terms than the one Kuhn and Hanson adopted. The reader should

realize, however, that neither Kuhn nor Hanson have especially the doctrines of

SR in mind; one could claim that their true aim is more ‘foundational’ in the

philosophy of science: they (especially Kuhn) challenge the realists’

understanding of scientific practice itself, i.e. the terms and the setting of the

realism debate, not any specific realist thesis or argument alone. As such, their

challenge addresses not the realists’ understanding of the facts, but what

actually counts as a ‘fact’ in the first place. I shall have occasion to return to

this approach in section 5, when I offer my own conclusions. At any rate, as

Boyd (2002) himself notes, this line of argument easily fits in with

anthropological relativism and social constructivism, neither of which addresses

the specific arguments in favour or against SR per se (NMA or PMI).

However, it is clear the detailed historical case-studies described by Devitt

and Psillos should be required anyway for the defence and explication of SR

doctrines, irrespective of PMI, since SR itself is put forward as an empirical

hypothesis to explain the success of science: how else is one supposed to test

realism if not by applying it in the analysis of actual historical episodes? As an

answer to PMI, though, realist historiography seems quite weak: case-studies of

historical events, realistically reconstructed, do not seem to have any special

90

relevance to the anti-realist arguments per se; one could take them as realist

‘just-so’ stories, as ‘retrospective realist’ exercises, and offer anti-realist

construals instead.62 Moreover,

the conservative policy of a retreat to believing only the least common denominator

across the history of science seems perverse, since in many cases our current

theories seem strongest precisely where they diverge from their predecessors.

(Lipton 2005: 1266)

In a sense, even the interest to undertake such case-studies would seem to

presuppose SR itself, rather than the opposite: uncovering formal and/or

conceptual similarities and continuities between past and present theories, on

the level of theoretical invariants and unobservable entities, can certainly not

dispel any serious anti-realist doubts on its own. Perhaps not even ‘success’

should be taken as narrowly as both realism and anti-realism take it: it is

perfectly conceivable that the major criteria for a theory’s successfulness are

not (or not only) empirical/observational; they may very well have more to do

with the cultural and the ideological background of the times than actual

empirical confirmations.63

62 ‘Retrospective realism’ comes from Andrew Pickering; see Nelson (1994), note 7. On the artificiality of realist stories, see Kukla (1996a), Nelson (1994), Stanford (2000), Psillos (2001b) and especially Stanford (2003). 63 Indeed, this position (specifically, that the criteria have been —and are— mostly aesthetic and pragmatic) has already been well-argued by McAllister (1993): “Philosophers who debate the plausibility of realism —realists as well as antirealists— frequently assume without second thought that theories which were in history counted as successful were judged to be so on the criteria reconstructed and prescribed by today’s philosophers of science. In particular, since most of the participants to the current debate concur in considering degree of observational success to be one of the most notable properties of a theory, they tend to assume that scientists have judged theories —and judged certain theories to be successful— by checking their degrees of observational success.” (ibid: 206)

91

Moreover, it is not entirely clear whether Psillos realises the full implications

of the move he is suggesting: Psillos’s and Devitt’s programmatic claims seem

to compromise SR as a global doctrine about science. As Mäki (2005) notes,

this localization turns scientific realism

into a more selective account sensitive to the nature and performance of particular

theories or disciplines. This has resulted in a shrinking scope for realism: those

parts of science which do not fit with some canons of realism are being excluded

and given over to anti-realist interpretations. (Mäki 2005: 231)

In other words, if one should go local and try out the ‘divide et impera’ move

(Psillos 1999: 108) on past and current scientific theories, why not do the same

for SR itself? In other words, why go for such strong commitments as the

metaphysical and the epistemic theses, if we can be selective in our ontological

commitments, and focus only on specific domains of science (e.g. accept a

realist construal for biology and astrophysics, but reject one for high-energy

physics and cosmology)? Why not be equally selective in our

ideological/metaphysical commitments too?64

Psillos’s move, admittedly a reasonable one, and consistent with his a priori

philosophical commitments for a global conception of SR, would seem to

deflect successfully the PMI-argument’s power against the epistemic thesis of

SR, at the cost of watering down the metaphysical thesis: the realist’s actual

ontological commitments turn out to be so case-sensitive and historicised, 64 See Fine (1991), who argues against the coherence of even such a position, and the response of McMullin (1991). That the standard methods for defending SR are not appropriate for the whole of science (e.g. historical science seems to be an exception) has been argued by Turner (2004), and his argument has been evaluated favourably by Carman (2005). Of relevance here is also the distinction of wholesale/retail arguments for SR in Callender & Magnus (2004). I come back to this distinction in the final section of this chapter.

92

epistemically and ontologically so dependent on the present understanding of

science, that they no longer seem strong enough to sustain a general,

metaphysical, and supposedly time-independent doctrine such as SR; in

practice, taking theories ‘at face value’ turns out to be either useless or trivial.

(What would such a contextualized realist be in general a realist about?) This is

a serious problem for the referential claims of SR, since there can be no

referential invariance of the supposed ‘natural kinds’ retained through theory-

change, if there are no entities that ‘stay put’, so to speak.65

Yet Mäki (2005) for one, while accepting all anti-realist criticisms (ibid:

235) as well as the above realist programmatic aims (ibid: 236), still does not

give up on the global scope of realism (or anti-realism). Strange as it may

sound, Mäki himself puts forward ‘doubly local scientific realism’ in order “to

make the concept of realism itself sensitive to the specific characteristics of

various branches of science” and in this way contribute “to a sort of re-

globalization”, “to re-globalize realism by going local in a more penetrating

manner than has been done by selective realists” (ibid: 232). At least prima

65 Carrier (1991) and (1993) rejects the No Miracle Argument in favour of an alternative construal which avoids commitment to any specific ‘natural kind’-account of referential invariance: he suggests ‘kind-structure’, the ‘underlying structure of likeness and difference in nature’ (1991: 34) as the non-referential element retained through theory-change. In any case, referential invariance, if developed for the purpose of countering PMI, seems to have “its epistemic priorities all wrong. For, we know far less about reference … than we know about what exists. In light of this, the rational procedure is to let our view of what exists guide our theories of reference rather than let our theories of reference determine what exists” (Devitt 2005: 785-6). For “interesting recent arguments to the conclusion that the realist is playing an illegitimate semantic game in trying to salvage her realism by tailoring theories of reference to ensure referential invariance” (Saatsi 2005: 518, note 9), see Bishop (2003), Bishop & Stich (1998), Douven & Van Brakel (1995), McLeish (2005), Saatsi (2005), and Stanford (2003a) and (2003b). In any case, it is not unusual for scientific realists themselves to criticise each other exactly on this point, e.g. Psillos (1997) criticises Kitcher (1993). In addition, it is not clear how the referential demands of scientific realism tally with the naturalistic approach adopted by realists —after all, the ‘arch-naturalist’ W. V. Quine defended ‘referential indeterminacy’ and the ‘inscrutability of reference’! See McGowan (1999).

93

facie, ‘doubly local scientific realism’ seems to be the ultimate, so to speak, in

watered-down realism and anti-realism:

Doubly local scientific realism (and antirealism): Some scientific disciplines or

their parts at a particular time invite a realist1 or realist2 or realist3 or …

interpretation while others may invite an antirealist1 or antirealist2 or …

interpretation. In other words, the adherence to realism is a function of kinds of

scientific units [Ri] and of kinds of realism [xi]: R1(x1), R2 (x2), …, ARk(xk), …

(ibid: 233)

What’s more, Mäki’s SR does not invoke any of the familiar arguments from

success, pessimistic meta-induction, or underdetermination (ibid: 235); he does

not take approximate truth or reference as the core concerns of SR (ibid: 236)66;

he talks of standard SR as defined by “features that are characteristic of the

physics-based attitudes that have dominated the debates over scientific realism”

and restricted by uses which are “outcomes of conversational trajectories that

no community of philosophers collectively and deliberately decided to follow”

(ibid).

These are impressive claims; however, even if accurate, one might wonder

that if this is what realism should become, what exactly is there then to object

about, in global or general terms. Surely, such conceptual flexibility in the

understanding of SR cannot sustain any global or general anti-realist argument

–or a realist one, either. Mäki has foreseen such an objection:

66 “[R]ather they are derivative concerns. I take realism to be primarily an ontological doctrine. Semantics and epistemology are important but not constitutive of the core concept of realism” (Mäki 2005: 236).

94

My first response to this objection would be the reminder that there is a sense in

which certain popular formulations of realism (and antirealism) already are

discipline-relative: realism has characteristically been formulated with one

discipline, usually physics or its representative parts, implicitly or explicitly in

mind. Thus, those popular formulations are themselves already relative to

discipline. My second response is a concession. The formulation of local versions

of realism R1(x1), R2 (x2), …, should be guided by some general realist background

intuitions that are not up for grabs (it is another, and perhaps disputable, matter

what exactly those intuitions are). So, there is a sense in which some general notion

of realism precedes – and constrains – the formulation of its specific versions. The

challenge would be to identify the generic realism that all such realisms should

share to qualify as realisms. The good news is that the endeavour of formulating

those local versions of realism may inform us about the proper formulation of a

generic realism that is able to accommodate them all. (ibid: 235)

It is too early to judge matters over this ‘new-generation’ SR: obviously, at

present this watered-down kind of SR lacks the appropriate vocabulary to

provide the argumentative structure one would like to see and evaluate. It

should be noted, nevertheless, that when all has been said and done, standard

SR has been more or less dissolved: the realism/anti-realism debate has not had

a straightforward resolution, but has been replaced by the search for new

programmatic aims, new projects and new concerns. That one should no longer

go uncritically for the standard global conception of traditional SR with its

associated construal of truth, success and explanation, is a fact which cannot

be emphasized enough.

95

Arguably, the dissolution of standard SR is a (direct or indirect) consequence

of PMI. So what do the above mean for PMI, then? Perhaps PMI itself should

also shed its global status and be understood in a very specific manner, tailored

for a specific philosophical target, namely the contention that an ‘empirically

successful’ theory should be deemed ‘probably approximately true’. This

connection of success and truth has been at the heart of the realist's intuition of

the No Miracle Argument (the best explanation of success of science is the

approximate truth of its theories). Perhaps PMI should be seen primarily as an

argument developed to undermine NMA, and should be appreciated solely in

that context, and not beyond it.

In any case, the jury is still out on the pessimistic meta-induction, and it is

not clear what the verdict is going to be; indeed, it is becoming increasingly

apparent that the grounds for this sort of anti-realist argument are not as strong

as it first appeared, and even the cogency of the whole reasoning might be

disputed: the argument itself may turn out to be actually fallacious67:

The series of theories [in Laudan’s examples] is not a set with independent

members from which we have taken a random sample, for the series is the output of

a process designed precisely to convert false theories into true ones. In this context,

the fact that past theories have been false does not provide a reason to believe future

ones will be false as well. Indeed, one might even argue that the history of science

to which the pessimistic induction appeals points in the opposite direction. …

Science is in the business of learning from its mistakes, and finding out what does

not work may be an indispensable guide to finding out what does work. So one

might go so far as to argue that we have more reason to believe some of our current

67 See Callender & Magnus (2004), Lange (2002), Lewis (2001), Lipton (2005).

96

theories in light of our extensive knowledge about how various earlier alternatives

to them have failed, than we would if, by some miracle, the current theory had been

the first one in its domain, a theory without a history. (Lipton 2005: 1266-7)

In other words, the ‘inductive base’ of the meta-induction, the time-dependence

of the ‘inductive base’, even the statistical conceptions implicit in its

formulation, may all be doubted (Lewis 2001).

Consequently, at this point things get rather more technical and conceptually

intricate than usual, in relation both to the statistical treatment (e.g. Douven

2005) and the historiographical details of the relevant episodes (e.g. Achinstein

2002). Contentious claims about truth-likeness, Bayesianism, formal theories of

confirmation, formal notions of approximate truth, even the status of normative

epistemology seem to be called for to keep the debate going.68 For this reason,

one must consider Worrall’s suggestion that both NMA and PMI might better

be called “considerations” for and against SR, and not arguments per se

(Worrall 1989: 101). Indeed, such suggestions motivate further opposition to

realism, though the alternatives have not been any less intricate or metaphysical

or less strongly committed to elaborate a priori distinctions (e.g. between

what’s observable and what’s unobservable, and between acceptance of a

theory and belief, both of which can be found in van Fraassen’s ‘Constructive

68 This is especially apparent in the review symposium of Psillos (1999) in Metascience: while examining NMA and PMI, the reviewers call for (new or better developed) formal accounts of confirmation (Redhead 2001) or truth-likeness (Bueno 2001), and only one of them (Lipton 2001) seems to doubt the cogency of the whole enterprise (on the grounds that realism as such adds nothing to the evidence for the truth of scientific theories which has not been obtained already by the scientists themselves; instead, it presupposes it). For appeals to naturalistic epistemology, see Devitt (1991) and (2005): 787; for Bayesianism, Douven (2005); for truth, Niiniluoto (1999); Smith (1998); Weston (1992). All of these appeals are contestable, of course, and have been contested. It certainly seems ironic, though –albeit neither unscientific nor unreasonable mind– that supposedly naturalistically-conducted debates (over how scientific evidence should be understood — realistically or instrumentally) turn out to harbour seemingly irresolvable, a priori contrasts very close to the surface.

97

Empiricism’). All the parties seem to disagree over what counts as evidence,

what is the right methodology to pursue the matter, and how much weight

should be accorded to explanatory reasoning. Perhaps the disagreements cannot

be resolved by the import of evidence and further argument: “[N]o

considerations of fact or logic … can –or should– persuade proponents of either

side [of the debate] to switch … [so that] the differences between realists and

antirealists are irreconcilable” (Kukla 1998: 153-4).69 Arthur Fine’s ‘Natural

Ontological Attitude’ seems to rest on a similar motivation. I will now turn to a

brief examination of these radical alternatives.

2.4 Weakening Strategies and Alternatives to SR

The lack of clarity, the tensions and the inconclusiveness implicit in the above

have certainly had an impact on the proponents of both SR and anti-realism.

Both realism and anti-realism have fragmented: some realists have tried to

weaken realism, while non-realists themselves have opted for entirely new,

alternative positions. Structural realism, entity-realism and Putnam's so-called

‘internal realism’ (Putnam 1981, 1983a), all belong to the camp of weakening

69 Kukla (1998) and Niiniluoto (1999) seem both to have been written in the wake of similar realizations. At least Kukla admits as much (quotation in the text; also, just before evaluating the underdetermination argument: “It’s a pretty close repetition of what I’ve had to say about dozens of arguments on both sides of the realism debate: they are question-begging and redundant. I apologise for being tedious. But my material demands it”, Kukla 1998: 89) and tries to provide reasons for an epistemology which actually accepts such ‘irreproachable irreconcilabilities’; Niiniluoto, on the other hand, acknowledges the tension, but tries obliviously to carry on the debate as if it’s business as usual: he offers formal accounts of concepts, puts forward an ontological picture, and avoids arguing in any depth for either! Clearly, under these conditions, there is no need to resolve the debate: one may simply offer pictures and accounts of already decided (a priori) realist views and beliefs, label them ‘critical’ and go on to carry implications (arguably, this describes quite well how SR has been received in IR). Thirty years ago, one might have proceeded in this cavalier manner without raising comments (Bhaskar ([1975] 2008) is a clear example of this attitude); today, however, one may legitimately talk of ennui and déjà vu, and abandon hopes for any real progress coming out of the realism debate; see Callender & Magnus (2004). Similar realizations can be found as far back as in Levin (1984) and Murphy (1990), and as currently as in Callender & Huggett (2011) and Lipton (2005).

98

strategies (though Putnam, as usual, is not entirely easy to place exactly).

Worrall’s epistemic structural realism (Worrall 1989) and the later ontic

structural realism proposed by French and Ladyman (French & Ladyman

2003), Bas van Fraassen's constructive empiricism (van Fraassen 1980) and

Arthur Fine’s ‘Natural Ontological Attitude' or NOA (Fine 1986a), belong to

the alternatives. Since these positions are quite elaborate philosophical

viewpoints that engage fully with all of realism’s philosophical commitments,

in order to do them full justice I would have to approach them as the stand-

alone positions they are meant to be and to develop them extensively and

independently of the debate on SR. However, this is beyond the scope of the

present work, and as such my exposition on them will be rather brief. Every

position above has something going for it, but they all differ, like entity-realism

does, from the core commitments of SR in their own specific and often unique

ways70.

Structural Realism (defended e.g. by Worrall 1989) holds onto the

mathematical structure of a theory, namely it claims that scientific knowledge is

restricted only to these structural aspects of reality which are encapsulated in

mathematical structure. Therefore, one may note that the truly radical aspect of

this version of structural realism is the attention it draws to the fact that today’s

scientific theories are also (and indispensably so) intricately structured

mathematical theories, and not a mere collection of physical concepts. This fact

had not been appreciated enough in the realist literature; also actual

experimental practice had been underestimated before entity-realism’s call for

attention to it. 70 The manner in which entity-realism (the position that gives up on theory-realism in favour of belief in the reality of the entities which experiments take to be part of the causal web of the world) differs from SR was referred to in section 3 of the present chapter.

99

In other words, Worrall’s epistemic Structural Realism gives up on its

specific ontological-conceptual commitments (the intrinsic nature of the entities

themselves)71 and tries to combine the lessons from the pessimistic meta-

induction, which was induced by anti-realists, and the no-miracle arguments

that I discussed above. In that respect Ruetsche notes that:

The Structural Realist takes both lessons to heart by espousing a realism about the

form of structure of our best theories, a realism which explains their empirical

success; these forms or structures inoculate themselves against pessimistic

inductions by carrying over to successor theories. (Ruetche 2011: 347)

Ladyman questioned Worrall’s epistemic Structural Realism because of the

latter’s failure to address the issue of ontological discontinuity that has

dominated the debate between realists and antirealists. Ladyman (1998)

proposed a radical form of ontic structural realism, which challenged the

metaphysics and the semantics of scientific realism and claimed that structures

are the only things that exist. Yet, this form of structural realism does not allow

for unobservable entities and, as a consequence, dismisses the role that the

discovery of these entities might have in the change of structures. The

discussion regarding the content of structural realism and its foundations is still

commencing, since structural realism aspires to offer a solution for overcoming

the earlier dilemmas that have been dominating the discussion between realists

and anti-realists (or empiricists). In his attempt to interpret quantum field theory 71 For exposition, see Bueno (2001), Chacravartty (2004), Ladyman (1998), and Worrall (1989) (perhaps Carrier 1991 is also relevant here). For criticism, see Cao (2003a, 2003b), McArthur (2003), Psillos (1995), (1999), chapter 7, (2001a), Redhead (2001) and (2007). For the latest defence, see Worrall (2007). Redhead (2001)’s comments are especially useful for understanding the meaning of the term ‘structure’ (a task already undertaken in current philosophy of mathematics, e.g. in Shapiro 1997).

100

(Cao 2003a, 2003b), Cao develops a kind of constructive structural realism vis-

a-vis the development of Quantum Chromodynamics (QCD) (Cao 2010).72

Dawid develops what he calls ‘consent structural realism’ in order to deal with

the fundamental challenges posed to ontology by string theory (Dawid 2007),73

which dominates our efforts to unify all natural forces (Weingard [1988] 2001).

Finally, Ruetsche (2002a, 2002b, 2011) argues in favour of taking “coalesced

structures seriously, notwithstanding the role idealizations play in constituting

those structures” (Ruetsche 2011: 349). This is because of the strong possibility

that those “structures might persist in future theories, a possibility lent some

plausibility by the role played by renormalization, group techniques, devised to

accommodate those structures, in shaping physicists’ quantum field theories”

(ibid: 349). In other words, Ruetsche challenges previous concepts about the

notion of structure and defies previous forms of scientific and structural

realism.

Putnam’s internal realism (Putnam 1981) transforms the concept of truth

itself, proposing a perspectivalism instead of the external-world correspondence

built into realism itself; while NOA gives up on both realism and anti-realism,

charging them both with unreflecting fixation on certain unrealistic,

unreasonable ‘pictures’ of science, and advising replacing theoretical 72 By examining the history of the evolvement of QCD and in particular the interaction between theoretical conceptualizations and experimental “observations” of quarks and gluons in the events of naked charmed particles and three jets, Cao proposes a more sophisticated form of structural realism, namely constructive structural realism. The basic assumptions underlying his theory are: “i) the physical world consists of entities that are all structured and/or involved in larger structures; ii) entities of any kind can be approached through their internal and external structural properties and relations that are epistemically accessible to us.” (Cao 2010: 6) 73 The consent structural realism proposed by Dawid (2007) is a more modest form of structural realism and is based on the assumptions that (a) one can accumulate knowledge about structures and that (b) one can gradually approach the one true structure implied by string theory. This form of structural realism attempts to do justice to the observation of real entities and the role they play in structuring our theories. Furthermore, this form of structural realism, also, takes into account the structures through which one accesses knowledge, the use of which can make the world observable to us. According to Dawid: “Thus emerges an intricate compound of mutual dependence between observational and structural aspects of reality” (Dawid 2007: 41).

101

commitment with a non-special (‘natural’) attitude toward science — one that is

minimal, deflationary (‘quietist’) and explicitly local. It advises the careful

examination of particular scientific claims and procedures, and cautions against

any general interpretive agenda toward science, in a rather anti-philosophical,

Wittgensteinian spirit. It certainly goes against the spirit of the entire

realism/anti-realism debate as it has traditionally been conducted — a fact

which, of course, has been noted (and roundly criticised) by scientific realists74.

Van Fraassen's constructive empiricism, by contrast, redefines anti-realism

itself, while trying to honour the core intuitions of the empiricist, anti-realist

position, in a way that is fitting for our times. In this spirit, van Fraassen

replaces empiricism’s distinction in terms of observational and theoretical terms

and predicates with one between observable and unobservable entities. Further,

his constructive empiricism takes empirical adequacy (not truth) as the goal of

science. As a result, when it accepts a theory, it accepts it as ‘empirically

adequate’, i.e. in an instrumentalist-like manner. This involves commitment to

working within the framework of the theory but eschewing belief in its literal

truth. At least prima facie, both distinctions —observable/unobservable and

belief/commitment— seem to fly in the face of the realist attitude toward

scientific theory and practice.75

Putnam’s views have been out of the debates for quite a long while now.

Putnam himself soon gave up on them and turned his back on the realism/anti-

realism debate in order to pursue other, less ‘metaphysical’, philosophical

interests. Structural Realism, as was manifested in the brief discussion above, 74 For criticism, see Kukla (1994b), Musgrave (1989), Psillos (1999), chapter 10. 75 For criticism and defence, see Alspector-Kelly (2001), (2004), Fine (2001), Giere (2005), Gutting (1983), Kukla (1995), (1998), chapter 8, Leeds (1994), McMichael (1985), Musgrave (1985), O’Leary-Hawthorne (1994), Psillos (1999), chapter 9, (2000a), (2000b), Rosen (1994), Teller (2001), van Fraassen (1994).

102

has drawn a lot of attention and has developed into a position with its own

associated literature of sustained criticism. Arthur Fine’s and Bas van

Fraassen’s positions have grown complex enough to deserve a whole chapter in

themselves. Both are the subjects of ongoing discussion and their associated

literature has grown so vast that a more detailed investigation of their work is

beyond the scope of this thesis.

Nonetheless, I have already reached the limits of standard SR, the specific

rationale of which should by now be clear.

2.5 Criticism and Conclusion

What ought to be clear by now is that standard SR can no longer be accepted in

its entirety as it is. SR has to be refashioned, in order to accommodate the

empirical failure of otherwise successful theories, as well as the discontinuity of

conceptual and ontological commitment, so evident in the history of science.

Realists have to be selective in what they are realists about: the literal reading

of entire theories (‘taking theories at face value’) cannot be sustained as it is.

All this was discussed in the analysis of the implications of the PMI-

considerations: the global, uncritical version of SR just doesn’t stand up to

historical scrutiny, either as an explanatory device for the success of scientific

theories, or as a philosophical interpretative approach to science. Indeed, what I

ended up with was, at best, Mäki’s ‘doubly local scientific realism’ or

something close to it:

In seeking to be informed about the sciences, doubly local scientific realism

subscribes to principles such as the following: the applicability of realism to a unit

103

of science is a matter that has to be decided locally, case by case; the information

needed for such decisions has to be acquired from local sources, by way of detailed

empirical examination of such units; the identification of the proper units of science

and the proper contents of realism is a locally interactive and empirical matter; any

claim about science more generally or more broadly has to be based on local

philosophical inquiries, in a bottom-up matter. (Mäki 2005: 234)

Therefore, in the process of assessing SR, one is to conclude that the truly

important level of the realist’s philosophical analysis lies at that of the detailed

mechanisms and the specialized theoretical elements of conceptually intricate,

formally sophisticated, fully mathematical, scientific theories. As I elaborated

in section 3, the methodological and the technical complexities of this level can

hardly support a grand-theory type of narrative76 such as standard SR. They

certainly don’t point directly to any kind of a realist interpretation, either. On

the contrary, as has been repeatedly argued by post-Kuhnian approaches to the

philosophy of science77, such a ‘piecemealist’78 approach would take one

directly out of the context of the realism debate and into examining the manner

in which the context itself has been set up. Indeed, our own suggestion shall be

that the context of the realism debate itself generates controversy, because of

the idealized conceptions of ‘science’ it harbours.

76 The word ‘narrative’ should not be taken as pointing implicitly to any postmodern type of critique, although there are affinities to postmodern critiques in what follows. However, ‘Postmodernism’, it should be kept in mind, is a very ambiguous term in the philosophy of science nowadays: it may encapsulate anyone from Quine (Murphy 1990) to Feyerabend (Preston 1998); mostly, the term comes up (unjustifiably) when discussing the Sociology of Scientific Knowledge programmes and similar stances (see Gal 2002 for a sober analysis of these in relation to realism). On the other hand, I take this occasion to explicitly distance myself from any deconstructionist versions of postmodernism: I have found them totally unhelpful in evaluating the realism debate; see, e.g., Parusnikova (1992). 77 Fuller (1989), Gal (2002), Galison (1988), Miller (1987), Roth (1996), Rouse (1981), (1987), (1991a), (1991b). Interesting realist treatments of these matters are Boyd (1992), (2002), Giere (1998), Kitcher (1993), Nola (1994), Radder (1993). 78 This term is borrowed from Rouse (1991): 141, who credits it in turn to Dudley Shapere.

104

I shall come back to these remarks shortly. For the time being, I would like

to prepare the ground by making three points.

First, the reader should note how the realists’ commitment to the complex

theoretical machinery of current science does seem to be, at least prima facie,

rather precarious. Current theories have many mathematically equivalent

formulations, and vary according to the purposes for which each is used. Each

one also employs a great range of incompatible models, according to the

experimental and calculational context of its use.79 In addition, engaging with

mathematically intricate theories requires a certain understanding of the

philosophical significance of the mathematical apparatus used; in other words,

significant problems in the philosophy of mathematics and applied science are

bound to crop up sooner or later. Indeed, it is not clear how realists can avoid

the debates over the so-called ‘indispensability argument’ in the philosophy of

mathematics (Colyvan 2001), which is implicitly involved in the reconstruction

of the ontological claims of a ‘true’ mathematical theory; cf. Resnik (1995).

Scientific realists systematically neglect the fact that, if empirical success is

taken to provide confirmation of a theory’s ontological hypotheses (the realist’s

assumption), then, since our current empirical theories are heavily

mathematised, this could be taken as confirmation of their mathematical

existence assumptions. One can see Leng (2005), who goes as far as to argue

“that it is scientific realists who should be most concerned about the issue of

Platonism and anti-Platonism in mathematics... The question of mathematical

ontology comes to the fore… once one considers our scientific theories,” (ibid: 79 This fact has given rise to the so-called ‘model theoretic’ approach to scientific theories; see the essays in Giere (1998) and Morgan & Morrison (1999). This approach should be distinguished from the formal, semantic view of scientific theories, according to which theories are construed as models of their linguistic formulations; see Chakravartty (2001) for a comprehensive review of the latter in relation to SR.

105

65). The same issue in anti-realism (van Fraassen’s version, to be specific) has

already received extensive treatment in the context of nominalism and

fictionalism in the philosophy of mathematics; see Bueno (1997) & (1999), and

Dicken (2006). In any case, van Fraasen himself was very well aware from

early on that, his own philosophy of science at least, had to face issues in the

philosophy of mathematics:

I do not really believe in abstract entities, which includes mathematical ones. Yet I

do not for a moment think that science should eschew the use of mathematics, nor

that logicians should, nor philosophers of science. I have not worked out a

nominalist philosophy of mathematics –my trying has not yet carried me thus far.

Yet I am clear that it would have to be a fictionalist account, legitimizing the use of

mathematics and all its intratheoretic distinctions in the course of that use,

unaffected by disbelief in the entities mathematical statements purport to be about.

(van Fraassen 1985: 303)

Second, and more troubling, is the fact that SR seems to ‘put all its eggs in one

basket’: everything that is philosophically important is taken to hang on the

metaphysical and epistemological status of the interpretation of theories, and

present theories at that; with current theories as points of reference, the realist

supposedly recognizes what has been retained and what not, employs them as

standards of comparison in order to circumscribe his/her commitments, and

restrict them according to the ‘story’ suggested by the realist reconstruction of

historical facts. Realists, then, give a picture of their activity as one of re-

description and re-interpretation both of scientific facts and scientific practice;

after having visited, so to speak, the scientific laboratory, they re-state, on their

106

terms, what the scientists tell them they found. It is no wonder, therefore, that

SR functions primarily as a template for understanding things, not merely

evaluating them; as such, it should be characterised as a guideline for re-

describing and re-conceptualising scientific practice in specific (realist) terms,

e.g, as

the doctrine that the sort of evidence that ordinarily counts in favour of the

acceptance of a scientific law or theory is, ordinarily, evidence for the (at least

approximate) truth of the law or theory as an account of the causal relations

obtaining between the entities quantified over in the law or theory in question.

(Boyd 1973: 1)

The right to impose such a picture on the ‘facts’, is, ultimately, what the debate

is all about. The realist commitments themselves, however, seem to have been

taken for granted in this endeavour all along: Why should ‘our current theories’

—our current points of reference— be given a realist interpretation in the first

place?

This, among other things, would imply that scientific inquiry always

converges on a single interpretation and that, for the philosopher, no

alternatives should be taken seriously when a scientific community has reached

a consensus over one theory. However, not only is it the case that the

convergence on a single interpretation of scientific inquiry is often debatable,

but it is also the case that philosophy has a lot to contribute to the debate when

examining alternatives. This is because unconceived alternatives may be of

significance still. As Stanford argues:

107

[I]f the historical evidence confirms that past practitioners have indeed routinely

failed to conceive of well-confirmed alternative hypotheses … that were

sufficiently serious as to be actually accepted by later scientific communities, then

we have every reason to believe that there are similar alternatives to our own

contemporary scientific theories that remain presently unconceived, even if we

cannot specify or describe them further. This challenge to scientific realism enjoys

several advantages over the traditional pessimistic [meta-]induction, but perhaps the

most important is that the problem of unconceived alternatives concerns the

theorists rather than the theories of past science: even if contemporary scientific

theories sometimes enjoy empirical successes arguably unprecedented in their

scope and character, this offers us no reason to suppose that today’s scientists are

any better at conceiving of the full range of theoretical possibilities confirmed by

this evidence than were the greatest scientific minds of the past. (Stanford 2006:

123, emphasis in the original)

Furthermore, our most accurate and fruitful scientific theories still await

plausible ‘realist’ readings. Take quantum mechanics, for example (Ruetsche

2002a, 2011; Wessels 1993), quantum field theory (Ruetsche 2002b), or even

the general theory of relativity (Belot 1996); both require extensive

reconstructive analysis in order to uncover exactly what (if anything!) a realist

commits oneself to. This is not merely a ‘philosophical’ problem: scientists

themselves do not know exactly what kind of a ‘realist’ claim the theories make

on reality (if any!), and for a full theory of quantum gravity (which we

presently lack), these interpretative issues are deemed to be important

(Callender & Huggett 2001). Nonetheless, this does not entail that a realist

construal of our current theories is indispensable or even preferable. On the

contrary, it is far more probable that our current (incompatible with each other)

108

theories, if realistically understood, are actually false. I shall have occasion to

revisit the full implications of this situation in Chapter 5, when I examine

Alexander Wendt’s ideas for a so-called ‘quantum social theory’.

Moreover, it is even less plausible that (a) one coherent picture of the natural

world in all its complexity arises once ‘the dust settles down’ and that (b) the

philosopher need only observe that picture. For instance, it is well-known that

our two most successful and ‘mature’ physical theories (general relativity

theory and quantum field theory) do not currently cohere with each other; they

are actually mutually inconsistent both in what each one predicts for the other’s

domain, and in the conceptual resources each one employs. Still, this has not

discouraged realists from taking mutually inconsistent theories at face value,

even if only in each one’s respective domain of application (Brown 1990).

Even if SR, as a global, ‘grand-theory’-type of philosophical picture, seems

to be no longer motivated (as noted, the picture has to be applied with care,

piecemeal, and after detailed acquaintance with the theory’s technical and

conceptual apparatus), the picture has to be there. Such passive acceptance,

though, is perplexing: Should it not be first the realist who should argue against

the passive acceptance of our current theories and confront them with

alternatives, in order to test their cogency? What better reason to take a theory

‘at face value’ than playing it against alternatives? The realist is after all

committed to the view that there is a coherent ontology underlying current

theory; not that an ontology corresponds to current epistemology (the epistemic

processes followed by scientists and the justifications and/or descriptions they

offer), which seems open to the vagaries of scientific practice. The realist’s

focus of interest should be ontology. One would not expect the realist merely to

109

trust in the results of a ‘consensus’, even if this consensus is a well-respected

scientific one. It may be wrong, however cogent the epistemic practice.

Certainly, as noted, the realist does not accept scientific claims uncritically; but

this is merely lip-service if no realist challenges them, either.

In addition (and this is my third point), one cannot help but notice how

organically connected the three dimensions of SR have shown themselves to be

in practice: once the semantic thesis has been settled, and one accepts that

scientific claims can be true and have existential purport, the metaphysical

thesis is ushered in to cash out what the semantics entails; truth and existence

are given a non-epistemic construal, which of course motivates our need to rely

on the epistemic thesis, securing thus our commitment to science as supportive

of belief in the truth of theories. But this in turn requires an even stronger

reading of the metaphysical thesis than the one I began with: a commitment to a

reality intelligible in a particular way, pictured as ontologically constrained by

referentially invariant entities or structures, natural kinds or mathematical

forms which somehow survive through theory-change and allow us to claim

that ‘science converges to the real’. Worrall, for example, unhesitatingly takes

it that, “in our more enlightened, post-Cartesian dualism-times” (Worrall 2007:

153)80, SR’s metaphysics should be more accurately formulated as:

There exists a structured reality of which the mind is a part; and, far from imposing

their own order on things, our mental operations are simply governed by the fixed

laws which describe the workings of Nature. (ibid: 153)

80 Attributing the formulation to Elie Zahar (2001: 86).

110

This metaphysical position has certain epistemological implications for

Worrall’s ‘structural scientific realism’ (hereafter: SSR, since ‘structural

realism’ was renamed ‘structural scientific realism’ in Worrall 2007):

Not only is this structured reality partially accessible to human discovery, it is

reasonable to believe that the successful theories in mature science – the unified

theories that explain the phenomena without ad hoc assumptions – have indeed

latched on, in some no doubt partial and approximate way, to that structured reality,

that they are, if you like, approximately true. (ibid: 154)81

The metaphysics, however, becomes more and more detailed: from step one,

the reality is ‘structured’, the mind has a specific ‘whole-part’ relation to it, and

scientific laws ‘describe the workings of Nature’. Although reasonable when

put in context (realism), for an anti-realist these claims are all far too

demanding, too strong, to take them seriously as defining the terms of the

realism debate. 82 This positive feedback, however — fleshing out the

metaphysics in order to accommodate the epistemology — might conceivably

go on and on, the realist position becoming more and more complicated with

every pass: indeed, the complexity of the current arguments and counter-

81 It is worth repeating at this point that the correspondence or semantic view of truth, which is assumed by SR as the way in which a theory mirrors reality is negated by Worrall (“But there is no reason why the way in which a theory mirrors reality should be the usual term-by-term mapping described by traditional semantics. Indeed, as I have remarked several times already, if we are talking about an epistemically accessible notion then it cannot be!” (Worrall 2007: 154)). Instead, “SSR [Structural Scientific Realism] in fact takes it that the mathematical structure of a theory may globally reflect reality without each of its components necessarily referring to a separate item of that reality. And it takes it that the indication that the theory does reflect reality is exactly the sort of predictive success from unified theories that motivates the No Miracle Argument” (ibid: 154). This position, which I personally find sophisticated and more close to the nature of the more advanced physical theories (e.g., general relativity, QM, QCD, quantum field theories, etc) is first and foremost due to Worrall’s profound reading of Henri Poincaré’s, Science and Hypothesis ([1905]1952), where Poincaré developed “a classic account of the No Miracle Argument” (ibid: 132). 82 Cf. the criticisms in Giere (1998).

111

arguments seem to demand it if the doctrines of SR are to be kept alive. Does

that mean that SR (or anti-SR, for that matter) does not even amount to a stable

position?

This last question brings me back to my previous remarks on the context of

the realism debate. One wonders whether something might have gone askew

with the setting of the realism debate itself. Could it be that both the realist and

the anti-realist have implicitly adopted a certain view of science (of the

theories, the laws, the mathematical apparatus, and the experimental practice)

which strongly biases the discussion in favour of the realist/anti-realist

philosophical divide, which naturally produces the controversy? It is entirely

possible that the realism debate has been founded on a set of dichotomies which

only produce questions, with neither the provisions nor the conceptual

resources to supply answers by themselves. Perhaps the realists should widen

their scope, and not merely ‘go local’ in their research habits, but also give up

their faith in the strong, ‘global’ claims entirely. Indeed, as it has already been

discussed, Arthur Fine has repeatedly argued for his NOA by construing the

realism debate as fixated on certain ‘pictures’ which are not naturally found in

scientific practice; thus, the realists and anti-realists beg the question even

before the debate begins. Admittedly, questions such as the above are difficult

ones to state precisely, let alone argue cogently about. However, they have

already been much debated in the literature (mostly over the coherence of the

reasoning behind the No Miracle Argument83). They concern deep issues of

83 Fine (1986a), (1986b), Laudan (1981). But see also Giere (1998), chapter 4.

112

philosophical method, namely naturalism, methodological pluralism, and the

place of metaphysics in philosophical (and scientific84) debate.

I would like to conclude my survey of SR in the philosophy of science by

suggesting that, when all is said and done, the realism debate is largely self-

sustained and self-generated; specifically, the features which produce the

realism/anti-realism controversy lie in the idealized conception of ‘science’

which both realists and anti-realists have uncritically adopted: it seems to me

that this conception imposes a gap and then tries to bridge it via (dis)ingenuous

technical tools and evasion of the truly relevant conceptual issues. This

conception of ‘ideal science’ consists in the pure, ‘mythological’ notion, the

one found in textbooks, research reports, and popularizations. This notion

adopts hastily the portrayal of scientists as going after truth and extra-human

realities, instead of parsing the scientific practice in more natural and weaker

terms, such as e.g. those of intelligibility and credibility. In the latter picture,

scientists try to make sense of the physical and social phenomena, producing

theories and ideas not in order to reveal the structure of a pristine, static and

ideal world, but in order to make those phenomena intelligible, accessible, and

open to tractable, credible and reliable, theoretical and mathematical treatments.

Hence, the strong, ambitious realist conception neglects outright the concrete

character of doing science, in favour of the ‘phenomena’ and the ‘pictures’

which scientists themselves provide in order to order and explain them. This

realist stance is not as innocent as it sounds, since in practice it amounts to a

strongly selective guideline for understanding science, its history and

sociological aspects, as well as how one should talk about theory: as a result,

84 In the context of methodological worries, see the excellent reconstructions of actual historical scientific debates in Chang (2001), Achinstein (2002) and Stein (1989).

113

the realists either dismiss the heterogeneity of science, its historicity and social

dimensions as irrelevant, or use these very same features to motivate the realist

quest for Truth and Nature, via a cleaner, more abstract and ultimately unreal

conception of science85. This is unfortunate, since in this way, SR seems to

underplay the radical character of truly novel developments which do not fit the

realist model (e.g. mature theories which are non-visualizable and essentially

mathematical — quantum mechanics — or extremely successful though made

up of cookbook-style techniques — quantum field theory).

It seems to us, therefore, that this restricted and selective attention to

scientific practice is unreasonable; by contrast, the kind of wider focus that was

hinted at above, which does not allow for easily drawn, sharp, unambiguous

distinctions, between the public, social, human factors (epistemology), and

‘Nature’ (ontology), provides for a more natural point to found an inter-

disciplinary (and consistently naturalistic) philosophy of science. In that case,

theoretical entities, both observable and unobservable ones, would work

essentially as part of the scientist’s mechanisms of making sense of Nature, not

in the sense of instruments or as (approximate) truth-claims, either, but as

distinctive and heterogeneous techniques of achieving intelligibility of the

phenomena, credibility (in the scientific community) and reliability (of any

technology that results). These aims impose their own constraints on scientific

practice, which may rhetorically (and innocuously) be glossed over as ‘the

quest for truth and the inner workings of nature’, but which have very little to

85 See, for example, Boyd (1992) and Kitcher (2001b); especially relevant to the latter are the criticisms of Roth (2003).

114

do with any specific ontological picture such as the one SR projects onto

science.86

The distinction between realism and anti-realism in that case would start to

blur: theories would be functioning neither as an instrument to calculate

experimental outcomes, nor as adequate representations of phenomena, but as

part of an interlocking and mutually supporting set of techniques which work

towards both of these goals at the same time: in other words, instrumental and

representational adequacy are always goals for a theory to aim for,

achievements, not the starting point of understanding theory. Indeed,

embodying both representation and calculation, a theory allows for the means

to further the goals of research itself, since, on the one hand, by pressing home

the instrumentalist aspect of theory as opposed to its representational function,

theory may propel itself forward by taking advantage of any development in the

calculational techniques; while, on the other hand, favouring realism and the

function of representation over instrumentalism, opens up new possibilities and

new theoretical ideas which the calculation will have to catch up to87.

This way of dividing the goals of theory into realistic portrayal and

calculational adequacy — instead of trying to characterize the properties of

theory in only one of these ways — seems to exclude the anti-realists’ fixations

as unmotivated as well: the anti-realist has to make room for both goals, not to

accommodate the one in terms of the other, as e.g. van Fraassen’s constructive

empiricist attempts to do, when he distinguishes between ‘commitment’ to a

86 Stein (1989) has offered a far better construal of this line of thought than one could afford here: Stein, examining historical episodes from science, gives a cogent argument for how neither realism nor instrumentalism make a difference in scientific practice; on the contrary, insistence on either soon outgrows its usefulness in producing successful theories. 87 Arguably, this is exactly what has happened to the frontiers of research in high-energy physics, such as string theory.

115

theory’s predictive function and ‘belief’ in the explanatory pictures it provides.

Blackburn puts the point eloquently:

The problem is that there is simply no difference between, for example, on the one

hand being animated by the kinetic theory of gases, confidently expecting events to

fall out in the light of its predictions, using it as a point of reference in predicting

and controlling the future, and on the other hand believing that gases are composed

of moving molecules. There is no difference between being animated by a theory

according to which there once existed living trilobites and believing that there once

existed living trilobites….What can we do but disdain the fake modesty: ‘I don’t

really believe in trilobites; it is just that I structure all my thoughts about the fossil

record by accepting that they existed’? (Blackburn 2002: 127-8)

As such, the realist’s concern of separating epistemology from ontology would

look largely derivative and unmotivated: it crops up after all the hard work of

securing intelligibility and credibility has been done, adding nothing to the

theories themselves (neither bolstering the evidence nor offering rational

grounds for taking them more seriously than the scientists themselves), it

cannot help in scientific debates in deciding to which theory one should

attribute truth, it cannot contribute in historiographical pursuits, and, finally, it

draws attention from all the interesting questions about science, such as: Are

scientific laws mere ‘generalizations’ or part of the scientist’s mechanisms of

making sense? How do models work? What is the role of mathematics? And so

on.

To all these questions, the realist reacts by invoking ‘truth’ and ‘capturing

reality’; I would suggest that such reactions undervalue by far the complexity of

116

the questions here: SR, if it is to stand for an intelligible and credible agenda for

answering them, has to do a lot of work first, work which any scientist would

probably find unproductive and unmotivated: it would have to go beyond

merely admitting that our current interests and perspective influence any

reconstruction we might offer; it would have to give a cogent description of

what exactly those interests are in SR’s case, and examine how it separates the

social from the natural, if it doesn’t simply take the separation for granted from

the beginning. The realist stories, which try to capture “what, in retrospect,

prove to be the pivotal points at which a successful direction of development

was initiated, whether or not their importance was appreciated at the time,

while suppressing the peripheral or abortive” (Leplin 1997a: 68), are no longer

enough, unless what the realist aims for, ultimately, is to partake in “all the

excitement of the realist claim that science engages in a cognitive activity that

pushes back the frontiers of ignorance and error” (Psillos 2000a: 708).

The full development of these thoughts would distract the discussion too

much. However, I would wish to register my dissatisfaction with the, in my

view, unreasonably strong, global claims of realism on the one hand (Worrall’s

SSR being the version of SR which I am in more sympathy with for it makes

more modest and more sophisticated claims), and my agreement with the

suggestion that SR rests on the uncritical adoption of an unreal conception of

science, on the other. This latter conception has not been constructed by

philosophers, although a lot of philosophy has been put into it. One might say

that this ‘mythological’ conception of science springs from scientists’ natural

reactions. Nevertheless, this does not absolve the philosophers from uncritically

adopting it instead of critically examining it.

117

The ‘mythological’ conception of science originates in physics and natural

science. Furthermore, it seems that it has also been guiding the self-

understanding of social science. IR, especially, does not seem to be an

exception; on the contrary, the Third Debate has been largely enacted with a

naïve conception of science at the background. For that reason, it would be

interesting to examine how SR has been received in the field of IR; specifically,

in the works of IR theorists Alexander Wendt and Colin Wight. In my view, the

strong commitment to the ‘mythological’ conception of science88 and the

simplistic treatment of SR in IR presenτ major shortcomings to IR meta-

theoretical considerations. This is going to be the focus of the next two

chapters. In chapter 3 I will deal with Bhaskar, the arch critical realist, on

whose work a huge part of IR meta-theory is mainly built upon. In chapter 4 I

will focus exclusively on the ways SR has been conceived, explored and

deployed within IR Theory.

88 Wendt (1999: 39): “[I am] a strong believer in science… I am a ‘positivist’”. Since Wendt was also a scientific realist at that point, ‘positivism’ points to his commitment, at the time, not to any specific positivist doctrines about unobservables, but to what I have referred to here as ‘mythological’ science in general.

118

3

SCIENTIFIC REALISM

AS DEVELOPED BY ROY BHASKAR

3.1 Introduction

There is a long debate concerning the question whether debates in the field of

philosophy of science have a bearing on the development of actual scientific

practice and theorizing. The philosophy of science is a philosophical discipline

which has the practice of science as its object. Its aim is not only to understand

the evolution of science and scientific change, as is the case of the history of

science, but also to provide an understanding of what science is and what can

count as scientific work. Philosophy of science, therefore, seeks not only to

understand the practice of science, but moreover to both justify and criticize it.

However, philosophy of science has often been regarded with contempt on

the part of working scientists, who often claim that the philosophers’ work is

ultimately irrelevant to the actual business of scientific practice.89 This claim

can be summarized in the following terms: science can and will continue on its

way to the extension and deepening of our knowledge and mastery of the

world, without needing either the guidance or the control of the philosophy of

science.

While this attitude might be considered dominant in the domain of the

natural sciences, it is far from universally accepted in the field of the social

89 For more on this view, which is held by many leading working scientists, look for instance at Steven Weinberg’s Dreams of a Final Theory (Weinberg 1994).

119

sciences. The latter, in their search for an understanding of the real

requirements of their practice both in themselves and in relation to the natural

sciences, have often and customarily turned to philosophy in general and to the

philosophy of science in particular. This specific interest in philosophy and

meta-theoretical investigation in general has often been regarded both as a

distinctive feature and as a flaw of the social sciences.

In the previous chapter, I have examined SR as a robust albeit problematic

philosophy of science with its own metaphysics, semantics and epistemology.

While the influence of SR in the field of the natural sciences remains uncertain,

its influence on the social sciences in general, and on the discipline of IR in

particular, can hardly be overestimated. Actually, one might claim that SR has

been the touchstone of the meta-theoretical debates that have dominated the

discipline during the last twenty years or so.90

How can this interest on the part of IR in the philosophy of science in

general, and in SR in particular, be accounted for? This question will be

discussed in detail in Chapter 5 of the thesis, where I will closely examine the

influence and the bearing of SR on IR. A brief, though schematic, answer can

be formulated in the following terms: the interest of IR theorists in SR is the

result of a revolt against a particular image, or some would say an ‘ideology’,

concerning what an IR theory should be like in order for it to be regarded as

scientific, or even meaningful. This attitude has been both acclaimed, as a sign

of scientific seriousness and theoretical rigour, and criticized, for being an

impediment to the production of actual substantive work.

90 This is true mainly due to the works of Alexander Wendt (1987, 1999), Heiki Patomäki (2002) Colin Wight (2006) and Milja Kurki (2008).

120

In the previous chapter, I have been able to testify to the richness and

complexity of the debates on SR, especially in the field of the natural sciences.

SR is far from being a uniform position; actually, it would be better to describe

it as a set of meta-theoretical presuppositions or ‘prejudices’ in the positive

Gadamerian sense of the term91, that seek to justify themselves in the light of

internal and external criticism. It is this feature of SR, as discussed in the

context of the philosophy of science, which provides it with its distinctive

dynamic and vibrant character.

Now, it should be stated from the very beginning that this kind of

complexity and lack of uniformity is absent in the debate on SR as discussed in

the context of IR. Actually, in the context of the meta-theoretical debates in the

field of IR, what counts as SR is the particular brand of SR developed by the

British philosopher Roy Bhaskar (Bhaskar’s version of SR is also known as

Critical Realism). While Bhaskar’s Critical Realism (CR) is in conformity with

91 As regards the notion of ‘prejudice’ in the positive, Gadamerian, sense of the term, see Hans-Georg Gadamer’s Truth and Method (1975). It is there where one reads: “Rather, a person trying to understand a text is prepared for it to tell him something. That is why a hermeneutically trained mind must be, from the start, sensitive to the text’s quality of the newness. But this kind of sensitivity involves neither ‘neutrality’ in the matter of the object nor the extinction of one’s self, but the conscious assimilation of one’s fore-meanings and prejudices. The important thing is to be aware of one’s own bias, so that the text may present itself in all its newness and thus be able to assert its own truth against one’s fore-meanings” (ibid: 238). Furthermore, cf. Gadamer in Conversation – Reflections and Commentary (Palmer 2001: 43-44), where Gadamer in a reply to a relevant question of Carsten Dutt, clarifies that: “[…] The idea that authority and tradition are something one can appeal to for validation is a pure misunderstanding. Whoever appeals to authority and tradition will have no authority. Period. The same thing goes for prejudgements. Anyone who simply appeals to prejudices is not someone you can talk with. Indeed, a person who is not ready to put his or her prejudices in question is also someone to whom there is no point in talking. One time, at the beginning of his career, Heidegger made use of the term Vorurteilsüberlegenheit [one’s superiority over prejudgement] as a corrective measure. This concept includes the capacity for conceding the correctness of the argument of the other person, and in cases where one does not know enough and one has trust in the better knowledge of the other person, one recognizes his or her ‘authority’ in the matter. “All our learning is based on this. Good judgement [a positive form of prejudgement] is a faculty one uses in taking action or in claiming to know something, and of course it is not something one acquires through book reading” (ibid: 44). The kernel of Gadamer’s argument lies, however, in Dutt’s next comment: “[…] Your true target is really a form of historical consciousness which has been the guiding consciousness of the modern humanities and social sciences, insofar as its insights into the historicity of its objects has been paired with a blindness to its own inextricable involvement in that history” (ibid: 44-45).

121

several views and intuitions of other versions of SR, such as the ones I

examined in the second chapter of this study, it departs from them in numerous

ways.

The dominance of Bhaskar’s philosophy in the meta-theoretical debates of

IR renders necessary a careful and scrupulous examination of his main

philosophical positions. For it is an undeniable fact that Bhaskar’s version of

SR has been introduced in IR without having been previously submitted to

critical scrutiny from a philosophical point of view, thus leading to an often

unreflective endorsement of his views. Thus, Bhaskar’s positions tend to be

unequivocally adopted as philosophical truths which an IR theorist should

either accept or reject, in which case the rejection of Bhaskar’s views would be

equated with a rejection either of philosophy altogether or, more simply, of

philosophy’s relevance in IR.

3.2 Two Dimensions of Scientific Practice

Bhaskar’s version of SR is developed in its most systematic form in his seminal

work A Realist Theory of Science (Bhaskar [1975] 2008).92 Bhaskar’s main

goal is to establish science as a social product directed towards an objective

world.

According to Bhaskar, science is a product of human activity. It is always

situated in a specific time and place in the context of the historical evolution of

human society. Science is a work of man as a social and historical being; some

would even say that science is man’s most distinctive or noblest work. Though

science has no existence independently of man, since it is made by man and the

92 Chris Brown claims that A Realist Theory of Science is “by common agreement, the ur-text of critical realism” (Brown 2007: 410).

122

locus of its existence is the human mind, it is directed upon objects that exist

wholly independently of man. The objects of science, the various things,

properties and laws of which we seek knowledge, exist in themselves in

complete independence of man’s efforts to acquire knowledge of them. They

can interact with men, they can change men and can be possibly changed by

men; but their being is distinct from the being of man.

Therefore, one should distinguish two aspects or dimensions of science: first,

the social dimension of science, which establishes science as a historical

product of the human society; second, the objective dimension of science,

which consists of the independent objects whose knowledge is the ultimate aim

of all scientific endeavour: the objective world or nature that science struggles

to render manifest. According to Bhaskar, two separate kinds of objects

correspond to these two dimensions of science: the ‘transitive’ and the

‘intransitive’ objects of science. The ‘transitive’ objects of knowledge

are the raw materials of science – the artificial objects fashioned into items of

knowledge by the science of the day. They include the antecedently established

facts and theories, paradigms and models, methods and techniques of inquiry

available to a particular scientific school or worker. (Bhaskar [1975] 2008: 21, cf.

194-195)

Two things are to be noted here. First, that science does not grow out of a

direct, non-mediated interaction with nature, but rather out of antecedent

scientific practice. There is no leap from non-science to science, no

“spontaneous production of knowledge” (ibid: 24), but rather a continuum of

transitive, temporary, fallible and socially constructed practices. The materials

123

of the scientist are no immediately given ‘data’, but previous scientific

knowledge and practice. As Bhaskar remarks, “[K]nowledge depends upon

knowledge-like antecedents” (Bhaskar [1975] 2008: 22). Second, the transitive

dimension of science consists of transitive objects of science. But in what sense

does it make sense to speak of transitive objects of science? Bhaskar’s position

is somewhat ambiguous on that point.

I will try to make it clear by taking an example of a transitive object of

knowledge, e.g. a particular scientific theory, for instance, Newtonian

mechanics. When one learns the theory, one learns two things: the theory itself

and the reality that the theory purports to describe. The theory itself is a social

and historical construct: it was produced by men in a definite historical

moment, in the context of a particular society, etc. The theory is formulated in a

particular natural or artificial language, and makes sense only in that language.

This theory is something to be learnt in itself and in that sense it is an object of

knowledge. But when one learns the theory, then, in the case that this theory is

true, one also learns something beyond the theory itself: one acquires

knowledge of the reality that the theory purports to describe. The same holds in

the case of a photograph, a picture, or a map. Suppose one gets hold of a map.

If the map is exact or correct, then the object of knowledge is, besides the map

itself, the spatial region that the map represents. There is, of course, an obvious

difference between a map and an actual spatial region: the map is of paper, 2-

dimensional, man-made; the region is physical, 3-dimensional, and not man-

made. But, in the case that the map is correct, there is also an identity: when

one looks at the map, in the case that one understands the map, one learns not

only the map, but the region the map represents. In that case it makes sense to

124

speak of two different objects of knowledge: the man-made, socially

constructed, transitive ones and the independent, objective, intransitive ones.

The existence of the ‘transitive’ and the ‘intransitive’ dimensions of science

is the real nexus of Bhaskar’s argument since, according to Bhaskar:

Any adequate philosophy of science must be capable of sustaining and reconciling

both aspects of science; that is, of showing how science, which is a transitive

process dependent upon antecedent knowledge and the efficient activity of men, has

intransitive objects, which depend upon neither. In other words, it must be capable

of sustaining both (1) the social character of science and (2) the independence from

science of the objects of scientific thought. (Bhaskar [1975] 2008: 24)

3.3 Two Rival Philosophies of Science

Bhaskar seeks, first of all, to establish the objectivity of the intransitive

dimension of science. In this effort, he sets his own philosophy of science

against two rival accounts concerning the nature of scientific objects93.

The first rival philosophy of science that Bhaskar takes into consideration is

classical empiricism whose main representative is David Hume. Classical

empiricism is usually interpreted as an epistemology, that is, as an account of

what can be known rather than as a philosophy of science. But, according to

Bhaskar, classical empiricism’s epistemology leads directly to a philosophy of

science and indirectly to a defective ontology. Classical empiricism’s main

thesis is that all knowledge comes from experience: experience exhausts all

knowledge of the world possible for us. What is apprehended in experience is

93 A short presentation of Bhaskar’s ontological scheme is to be found in the entry “Ontology” written by Bhaskar himself for The Blackwell Dictionary of Modern Social Thought (Bhaskar 2003: 442-443).

125

an individual or atomistic event, e.g. that x is the case. The world available to us

is nothing other than a series of events presented to us in the context of discrete

experiences. Then, what is the purpose of science, according to classical

empiricism? It is to establish connections between events, that is, to show that

an event of type A is constantly followed by an event of type B. Therefore, the

purpose of science is to establish constant conjunctions of events according to

Hume’s term, or regular sequences of events as Bhaskar preferably calls them.

The world is reduced to events and to the possible connections of those events.

According to Bhaskar’s imagery, the world for classical empiricism may be

viewed as a ‘surface’, that is, without any depth, with no underlying structure.

Events are like points on this surface linked to one another with discrete lines.

According to classical empiricism, the ideal of science would be to establish

links between all the points on this surface. It is important to note that this

surface exists only in the human mind because the events that make up this

surface exist solely as phenomena of experience. For classical empiricism,

trying to discover what lies behind those phenomena would only be an attempt

to return to some kind of dogmatic metaphysics.

The second rival philosophy of science that Bhaskar takes into consideration

is transcendental idealism, which is represented by Immanuel Kant and the neo-

Kantian philosophers. According to transcendental idealism, nature is not

reduced to atomistic events. Behind the events there is an order, a structure or,

to use Kant’s term, a lawfulness (Gesetzmäßigkeit). However, this structure is

not a property of the objects themselves given to experience. It is rather to be

understood as a product of the human mind, which in its cognitive activity

imposes an order upon phenomena. For transcendental idealism, knowledge is

126

not acquired through the passive state of mere observation. Knowledge is an

activity through which the human mind organizes the data of experience into

models of the world. The world ceases here to be seen as a surface. However,

the depth - the underlying structure - that the world is provided with is not to be

viewed as a property of the world itself, but rather as a product of human

cognitive activity.

Now, there are two ways to interpret the nature of this cognitive activity:

either as an activity that seeks knowledge as an end in itself (in the spirit of

Plato or Aristotle)94 or as an activity influenced by the various interests and

dependencies that shape human striving in the world in general (in the spirit of

Hegel or Habermas).95 But, if cognitive activity is shaped by particular interests

and viewpoints and if the structure of the world is something imposed upon the

world by cognitive activity, then it follows that for transcendental idealism,

properly interpreted, “structure becomes a function of human needs; it is denied

a place in the world of things” (Bhaskar [1975] 2008: 28).

The common element that connects both rival philosophies is their

commitment to a particular ontology that Bhaskar describes as empirical

realism. Empirical realism is the view according to which the principal

category upon which any account of reality should be founded is the category

of experience or experience ability. For both classical empiricism and

transcendental idealism, no knowledge can exist of what exceeds the limits of

94 On Aristotle’s analysis of knowledge as a natural human end in itself, see Aristotle, Metaphysics, Book I, 1. On his theory of theoretical knowledge as man’s highest good, see Aristotle, Nicomachean Ethics, Book X, 7-9. See also Richard Kraut (1991), especially Ch. 1 & Ch. 3 and, also, Thomas Nagel (1972). On Plato’s views, see T. H. Irwin (1995), Ch. 20, §§ 233-234. For a comparison between Aristotle’s and Plato’s views on the subject, see A. W. H. Adkins (1978). 95 On Hegel, see Neuhouser (2003), especially Ch. 4-6. On Habermas’ views on the subject, see Habermas (1973), (1984), (1987); Rehg (2009), Ch. 4 & Ch. 5.

127

possible experience. But, apart from the object of which there is experience, the

notion of experience presupposes necessarily the subject for which there is

experience.96

This necessary implication of subjectivity in the notion of experience

combined with the belief that all statements about reality should be limited to

the domain of what can be given in the context of a possible experience, lead

inescapably to the common predicament of modern philosophy: reality should

always be described in relation to the knowing subject or, in other words, no

reality can be thought of that is independent of man. This predicament is

illustrated in the positions of three of the most prominent modern philosophers

vis-à-vis the objectivity of the external world. For Berkeley, there is no external

world; reality exists only as an idea of the knowing subject.97 For Hume, the

reality of the external world is something that one cannot help but believe in,

however, for this belief one can have no rational grounds; thus, one is

inescapably led into scepticism.98 For Kant, the reality of the external world is

something which cannot be known by way of philosophical argument.

However, this world is, first, a world whose structure is imposed by human

cognitive activity: the world’s structure is not an essential property of the

world itself. Second, this world is only a small portion of the world at large; the

external world of whose reality we can have philosophically grounded

knowledge is the world in so far as it is capable of being given in the context of

96 It is worth mentioning here that the following argumentation, which is based on the concepts of subject and subjectivity, constitutes an original reconstruction of the Bhaskarian empirical realism and is not to be found as such in the Bhaskarian corpus. 97 On Berkeley’s immaterialism, see Winkler (1989); Pitcher (1977); Pappas (2000). 98 On Bhaskar’s interpretation of Hume, see Bhaskar [1975] 2008: 24-7, 37, 40-2). On Humean scepticism, see Fogelin (1994: 209-237).

128

a possible experience; of the world which exceeds the limits of possible

experience we can have no knowledge whatsoever.

Bhaskar remarks that the crux of this ontology that defines reality in relation

to the knowing subject is the concept of the ‘empirical world’, of the world that

is the object of a possible experience. A necessary consequence of empirical

realism, which as I have discussed, characterizes both classical empiricism and

transcendental idealism, is a denial of the intransitive dimension of science: the

world of which science seeks knowledge is not independent of man; on the

contrary, it can only be described in relation to man and his cognitive activity.

3.4 Transcendental Argumentation: From Philosophy of Science to

Ontology

Bhaskar sets himself to provide an alternative philosophy of science that fully

acknowledges the intransitivity of the world that science seeks to discover. He

calls this philosophy of science ‘transcendental realism’. For transcendental

realism, the objects of knowledge:

(…) are neither phenomena (empiricism) nor human constructs imposed upon then

phenomena (idealism), but real structures which endure and operate independently

of our knowledge, our experience and the conditions which allow us access to them.

(Bhaskar [1975] 2008: 25)

Bhaskar’s aim is to demonstrate a fact about the world: that it exists wholly

independently of human cognitive activity. In this sense, Bhaskar seeks to

establish an ontological claim. However, his point of departure is a questioning

that concerns the objects of scientific activity; in other words, he begins with an

129

investigation in the field of the philosophy of science. This blending of the

philosophy of science with ontology is manifest in the methodological strategy

that he follows in order to establish his thesis on the independent existence of

the objects of scientific research. He presents a transcendental argument, that is

to say, an argument whose aim is to demonstrate a truth about the world starting

from a questioning about what the world should be like for scientific activity to

be intelligible.

There is no necessity that there is scientific knowledge. We can easily think

of a world without man or any form of intelligent life. A world could have

existed that would have never become the object of scientific investigation. But

there is scientific knowledge or at least scientific research. Yet, this fact is not

necessary; its negation is possible. But given that it is an actual fact, it has to be

possible.99 For only what is possible can be a fact, for there are no impossible

facts. Now, if there is scientific knowledge, or at least scientific research, one

may ask: how is scientific research possible? In other words, what must be the

case for scientific research to be possible? Or, to use Kantian terminology, what

are the conditions of possibility of scientific research?

One may think of many types of answer to this question. For scientific

research to be possible, man himself should be in a certain way. Therefore,

there are certain anthropological or psychological conditions of possibility of

scientific research. Furthermore, society should be in a certain way. Thus, there

are sociological or historical conditions of possibility for scientific research. A

Habermasian or a neo-pragmatist like Richard Rorty would add that there 99 “It is not necessary that science occurs. But given that it does, it is necessary that the world is in a certain way. It is contingent that the world is such that science is possible. And, given that it is possible, it is contingent upon the satisfaction of certain social conditions that science in fact occurs. But given that science does or could occur, the world must be a certain way” (Bhaskar [1975] 2008: 29).

130

should exist a specific normative framework for science to be possible. So,

certain ethical conditions of possibility of scientific research should also be

recognized. But, what is the most important for scientific research to be

possible is that the world itself must be in a certain way. Therefore, there can

and must be recognized certain ontological conditions of possibility of

scientific research.

To sum up the argument so far: Bhaskar argues that there is no necessity for

the occurrence of science. Nevertheless, since science does occur, science is

possible. But for science to be possible, certain conditions have to be met.

There are many different kinds of conditions for the possibility of science;

moreover, there must be certain ontological conditions for this possibility.

Therefore, one might seek to discover some truths about what the world is like

through a research on the ontological conditions of possibility of scientific

research. This is what a transcendental argument is. It is an argument that

purports to establish a truth about the world through an investigation on the

conditions of possibility of cognitive activity at large or scientific activity in

particular.

Finally, it is worth mentioning that the notion of a transcendental argument

is completely missed by Andrew Collier in his Critical Realism: An

Introduction to Roy Bhaskar’s Philosophy (Collier 1994). According to him, a

transcendental argument is simply an argument that searches the conditions of

possibility of an actual phenomenon. Therefore, a transcendental argument

would explain what must be the case for an actual phenomenon to take place.

He says that the form of a transcendental argument is the following: “from

something that is actual, to a more fundamental ‘something’ that grounds its

131

possibility” (ibid: 20). Now, this is not what a transcendental argument is. This

is just what a simple explanatory argument is. Suppose it rains, and someone

asks “How is this possible?”. The argument that would explain the conditions

of possibility of raining would not be a transcendental argument. It would

simply be an explanatory argument, i.e. a simple explanation. A transcendental

argument is something extremely more specific than that. It is an argument that

demonstrates a truth about objective reality through an investigation on the

conditions of possibility of cognitive activity. Thus, Kant in the

‘Transcendental Deduction’ of his Critique of Pure Reason purports to establish

a truth concerning the structure of nature through an investigation on the

conditions of possibility of empirical knowledge. Therefore, through his

‘Transcendental Deduction’ he establishes that the subjective conditions of

possibility of experience are at the same time necessary laws of nature. He

summarizes his point in his famous ‘supreme principle of all synthetic

judgments’: “the conditions of the possibility of experience in general are at the

same time conditions of the possibility of the objects of experience, and on this

account have objective validity” (Kant 1999: 283).

3.5 The Intransitivity of Objects of Sense-Perception

After having established his methodological strategy of transcendental

argumentation, Bhaskar embarks upon establishing his account on the

intransitivity of the objects of scientific enquiry. He presents two different

transcendental arguments. The first one focuses on the conditions of possibility

of sense-perception, while the second one, which is by far the most important,

focuses on the conditions of possibility of experimental activity.

132

The transcendental argument that focuses on the case of sense-perception

runs as follows: what has to be established is that the object of sense-perception

exists wholly independently of the subjective act of perceiving or the sense-

perception itself as a mental entity. As Bhaskar claims, “the intelligibility of

sense-perception presupposes the intransitivity of the object perceived”

(Bhaskar [1975] 2008: 31). In order to establish this thesis, his argument is

based neither on a particular philosophy of mind nor on any kind of

metaphysical presuppositions. Instead, Bhaskar examines the conditions of

possibility of certain forms of our cognitive activity.

His argument, which he exposes in an extremely condensed (and rather

ambiguous) form, could be reformulated as a reductio ad absurdum:

• Suppose that the object perceived does not exist independently of sense-

perception itself;

• Then the object of sense-perception should be construed as a product of the

act of perceiving itself existing solely as a mental representation or idea.

• Now, it is an actual fact that men can perceive an object in different ways.

For instance, one can look at an object from a different angle and see it

differently, or one may acquire a certain information on the object which

leads one to perceive the object in a different way;

• But if the object perceived is a mental creation of the perceiving subject,

then this change in the way of perceiving the object cannot be interpreted as

a different perception of the same object.

• Instead, one would have to accept that what is really happening in the case

of a change in the way one perceives a thing is that the perceiving subject

133

creates two (or more) distinct mental representations through its perceptual

activity. One does not see the same object in different ways. On the contrary,

it should be interpreted as creating two different mental objects.

• This means that perceptual change, the idea that the same object is seen in

different ways, should rather be viewed as a mere illusion.

At this point comes Bhaskar’s transcendental claim that in order to warrant

the possibility of perceptual change, one should acknowledge the independent

existence of the object of perception.100 The aforementioned represents a move

from an examination of the conditions of intelligibility or possibility of a

practice of ours - in this case: change in the way we perceive a thing - to a

claim concerning the nature of the real world. In order to be able to make sense

of something that we do in the context of our cognitive activity, we are led to a

claim about external reality.

It is important to note the hypothetical character of Bhaskar’s argument: the

world must be interpreted as independent from our cognitive activity, if

perceptual change is to be understood as a case of different perceptual

representations of the same object, rather than a case of a creation of two

different mental objects. Yet, Bhaskar offers no argument to establish that

perceptual change should be understood in the first way rather than in the

second one. It seems as if one were left with a choice between two prima facie

equally plausible interpretations of our cognitive activity. If one accepts the

100 This codification of Bhaskar’s transcendental claim in which I make use of the reductio ad absurdum method is an original one for it is not to be found in this form in any of the philosopher’s or his own critics’ and commentators’ texts. This way of presenting it will hopefully help in ‘demystifying’ its inner structure and logic by making it first of all comprehensive. Some of the critics of his book ARTS believe that its author fails to ground his transcendental claim but arguably they are not right (cf. Caroline Whitbeck’s review of this book in Whitbeck 1977: 114-118).

134

first interpretation, one is forced to admit a particular account of the nature of

reality; the same holds in the event that one accepts the second interpretation.

Different interpretations of our cognitive capacity have different ontological

commitments. Then what is the reason to prefer one interpretation to the other?

Bhaskar’s account seems to be that the first interpretation is faithful to the way

working scientists understand their own activity, that it is “consistent with our

intuitions” (Bhaskar [1975] 2008: 33), while the second one would force

scientists to abandon crucial aspects of the image they have of their own work.

Based on his argument concerning the conditions of intelligibility of sense-

perception, Bhaskar draws a categorical distinction between two different kinds

of entity: experiences, which belong to the domain of the empirical, and objects

perceived, which belong to the domain of the actual. It was shown that for

classical empiricism the objects of scientific research are reduced to events and

sequences of events. Bhaskar’s transcendental argument purports to

demonstrate that even if one admits that the objects of scientific research are

events and sequences of events, those events should be categorically

distinguished from the experiences through which one has cognitive access to

them.

Before moving to the examination of Bhaskar’s argument that focuses on the

conditions of intelligibility of experimental activity, it is helpful to take note of

how Bhaskar’s position is situated vis-à-vis the other philosophers that he takes

into consideration. Contra Berkeley, he establishes that the objects of sense-

perception are categorically distinct from the acts of perceiving. Contra Hume,

he believes that the independent existence of the objects of sense-perception

can be demonstrated by way of philosophical argument, rather than be accepted

135

as a matter of mere faith. Further, Bhaskar is in agreement with Kant’s critique

of Hume’s vision of the order of reality. He agrees with Kant that the world

should not be viewed as a surface consisting of events and constant

conjunctions of events. The world and the objects of the world have an order,

an underlying structure. But contra Kant, he wishes to establish that this

structure is not a simple product of human cognitive activity, but a property of

the world itself. This claim is the demonstrandum of his famous transcendental

argument that analyzes the conditions of intelligibility of experimental activity:

the structured nature of the world and the intransitive character of this structure.

3.6 The Intransitivity of the World's Structure

Hume believed that a causal law is nothing other but a constant conjunction of

events. A law of nature is simply the general fact that an event of type A is

constantly followed by an event of type B. For Hume, this regular sequence of

events is simply what a natural law is. But, how can one establish the existence

of a particular natural law? The obvious answer would be that one could

establish it through scientific research. And what is the most distinctive feature

of scientific research? Hume’s answer is again obvious: experimental activity.

It is at this point where Bhaskar’s next transcendental argument begins.

Bhaskar makes the rather trivial remark that an experiment does not exist by

itself. An experiment is not a natural happening; instead, it is something

produced by men. The aim of an experiment is, initially, the co-production and

then, the observation of a specific regular sequence of events. If this particular

regular sequence of events occurred by itself, independently of any human

interference, there would be no need to make an experiment. What renders

136

experimental activity necessary is precisely the fact that the sequence of events

under question does not normally take place by itself. Therefore, one may say:

without human interference, the sequence of events whose occurrence a

scientific inquiry seeks to testify does not take place. Here comes the next step

in the argument:

(1) Following Hume, if one equates a causal law with a particular regular

sequence of events;

(2) if no causal law can be scientifically identified without a proper antecedent

experimental procedure;

(3) if the goal of an experimental procedure is the occurence of a particular

regular sequence of events;

(4) if the occurrence of the particular sequence of events cannot take place

without an experimental procedure;

(5) if in every experiment man acts as a causal agent, that is, if every

experimental procedure involves human activity;

(6) if what exists as a result of the activity of some object, some person, some

force or some law is a direct or indirect product or effect of that object,

person, force or law;

Then, it follows that:

(7) all causal laws identified by science are the direct or indirect products or

effects of man.101

This conclusion follows naturally from the premises: the regular sequence of

events produced by an experiment exists as a result of human activity; therefore

101 This ‘codification’ of Bhaskar’s argument is original since it can be found neither in any of his own works nor in his critics’ publications.

137

it is a direct or indirect effect of product of man. Furthermore, the regular

sequence of events produced by an experiment does not take place naturally. If

it did, there would be no need for an experiment. Therefore, it can only take

place after the occurrence of an experiment which means that it can only exist

as a result of human activity. Therefore, it is necessarily a direct or indirect

effect or product of man. Now, if a natural law were a sequence of events, then

the sequence of events produced by the occurrence of an experiment would be a

natural law. Therefore, the natural law would be a result of the conduction of

the experiment and, thus, should be considered as a direct or indirect effect of

man. Finally, if no natural law may be scientifically recognized without the

antecedent taking place of an experiment, then it follows that every natural law

that is recognized by science is the direct or indirect effect of man.

What one has here is again a case of a reductio ad absurdum. In order to

avoid the absurd conclusion that the laws that govern nature are made by man,

Bhaskar claims that Hume’s account of natural causation should be completely

abandoned in favour of the following two theses:

(a) A causal law is not the same thing as a regular sequence of events. “There

must be an ontological distinction between them” (Bhaskar [1975] 2008:

33).

(b) A causal law continues to operate independently from any experimental

activity, that is, “outside the contexts under which the sequence of events is

generated… Causal laws endure and continue to operate in their normal

way under conditions, which may be characterized as ‘open’, where no

constant conjunction or regular sequence of events is forthcoming”, given

138

the fact that “outside astronomy, closed systems, viz. systems in which

constant conjunctions occur, must be experimentally established” (ibid:

33).

Berkeley’s primordial mistake was to identify events and experiences. Now,

one can see that Hume’s primordial mistake was to identify regular sequences

of events and causal laws. If one accepts Hume’s account of natural causation,

one is forced to consider that causal laws are products of man, which is an

absurdity. To avoid this absurdity, one should admit that the causal laws of

nature are categorically distinct from the regular sequences of events that occur

by way of our experimental activity. A regular sequence of events is to be

understood as an instance or a manifestation of a certain causal law and not as

the causal law itself. Furthermore, in order to make sense of experimental

activity, one should admit that a causal law does not cease to operate outside

the context of an experiment. An experimental framework should rather be

construed as a particular instance of the operation of a causal law, in the context

of which a particular law of nature becomes “actually manifest and empirically

accessible to men” (ibid: 46).

What one has here is another case of Bhaskar’s transcendental

argumentation. Bhaskar moves from an investigation of the conditions of

intelligibility of an activity of ours, viz. experimental activity, to a claim

concerning the world itself. For experimental activity to be properly intelligible,

one should assume, contra Hume, that causal laws are not constant

conjunctions, that is, regular sequences of events. A causal law is something

over and above a regular sequence. But, what exactly is a causal law?

139

Analyzing the nature of natural causation is the next step in Bhaskar’s

argument.

Before moving to the analysis of Bhaskar’s account of causation, it is

important to sum up what Bhaskar has accomplished so far. Bhaskar has

presented two transcendental arguments, viz. two arguments that establish

propositions about what the world is like from an examination of the conditions

of possibility or intelligibility of two cognitive activities of ours: sense-

perception and experimentation. These two arguments allow him to establish

several ontological claims, namely that:

(1) events are categorically distinct from experiences;

(2) the objects of experience exist independently of experience;

(3) a causal law continues to operate outside the context of an experiment;

(4) a causal law “presupposes a ‘real something’ over and above and

independent” (ibid: 49) of a regular sequence of events.

3.7 From Natural Causation to Generative Mechanisms

Starting from an investigation on the philosophy of science and more

specifically on an investigation concerning the objects of scientific knowledge,

I have presented a preliminary account of Bhaskar’s ontology, on which I will

now elaborate further.

Bhaskar’s use of the concept of ‘ontology’ is of seminal importance for

construction of his theoretical framework. According to Bhaskar: “The answer

to the transcendental question ‘what must the world be like for science to be

possible?’ deserves the name of ontology” (ibid: 23). This means that for

Bhaskar ontology is the discipline that investigates a particular set of conditions

140

of the possibility of scientific knowledge: the conditions that have to do with

what the world must be like for science to be possible. Given Bhaskar’s use of

the term ontology, it becomes obvious that ontological questions can only be

approached by use of transcendental arguments. This agrees with Kant’s

conception of ontology. But to understand this point, one has to take a closer

look at Kant’s critical project.

Traditional metaphysics was divided into two parts: ‘general metaphysics’

and ‘special metaphysics’. General metaphysics investigated what it is to be a

being, viz. what it is for something to exist. After the 17th century, general

metaphysics took the name ‘ontology’, from the Greek word ‘ον’ which means

‘being’. Special metaphysics was divided into three parts: psychology, which

investigated what it is to be a rational subject in general, cosmology which

investigated what is to be an object in general, and theology which investigated

the ultimate cause of being in general.102

Now, Kant’s aim was the deconstruction of metaphysics as a theoretical

discipline in so far as metaphysics is an attempt to acquire knowledge through

reason of what cannot be an object of a possible experience. His critical project

was divided into two parts: a deconstructive part, named ‘Transcendental

Dialectic’, where the three branches of special metaphysics are shown to lead

inescapably into antinomies and incoherence and are first abandoned as

theoretical disciplines and then transformed into objects of moral investigation;

and a constructive part, named ‘Transcendental Analytic’, which does not

eliminate ontology or general metaphysics, but rather transforms them into an

investigation on the mind’s contribution to the constitution of the empirical

102 See Heidegger (1997: §§1-2).

141

world. This transformation of traditional ontology into an Analytic of the

Understanding is accomplished precisely through the use of transcendental

arguments. In this way Kant manages to demonstrate that nature’s most general

laws are but the most general categories of mind. The proof of this is the object

of his Transcendental Deduction and his Analytic of Principles. Three things

are to be noted here:

(1) Bhaskar recognizes transcendental argumentation as the sole

methodology for arriving at a philosophical ontology. Actually, he even

describes ontology as an answer to a ‘transcendental question’, as he calls it.

From this point of view, he can be considered a Kantian concerning his

conception of what is ontology. At the same time, one should notice that this

Kantian conception of what ontology is appears in conflict with his stricto

sensu ontology in itself, which is thoroughly, though not explicitly Aristotelian,

as I will explain later.

(2) While the aim of Kant’s philosophy is to wholly discredit any attempt to

transgress the limits of possible experience, Bhaskar positively affirms that

philosophical ontology must go beyond the limits of what is given in

experience, reflecting upon the ‘real basis’ of experienceable events.

(3) While Kant transforms ontology into an Analytic of Understanding in

which the mind’s a priori categories are shown to be nature’s constitutive most

universal laws, Bhaskar’s transcendental argumentation purports to prove the

existence of an ‘ontological realm’ wholly independent of the human cognitive

activity.

In order to provide more depth and thickness to his ontology, Bhaskar

elaborates on his account on natural causation.

142

Proposition (4) in section 6 above indicates that a causal law presupposes a

‘real something’ over and above a regular sequence of events. Through the

logical analysis of this proposition, one could arrive at a deeper understanding

of the Bhaskarian nature of causation. Moreover, one could comprehend the

latter in the following manner:

(1) There is something that is beyond events and regular sequences of events.

Contra Hume, the world is not a surface. Beyond events, there is an

underlying structure.

(2) This ‘something’ is real. Contra Kant, it is neither a product of the mind’s

cognitive activity, nor something imposed by man upon reality, nor a

figment of imagination.103

(3) This ‘something’ that subsists beyond the events is operative; it is

something active. 103 It seems as if, according to Bhaskar, causal laws for Kant are imposed by the mind upon phenomena. This is a complete misunderstanding of Kant’s theory of causation. According to Kant’s ‘Second Analogy of Experience’ of the Critique of Pure Reason, “all alterations occur in accordance with the law of the connection of cause and effect” (Kant 1999:304). This proposition, which is a general principle of nature in its formal sense, that is as of nature as the lawful totality of objects of possible experience, is demonstrated through a transcendental argument whose aim is to establish how an objective law of nature derives from a set of subjective a priori conditions of experience. It is in this sense that one might say that the general principle of causality described in the Second Analogy of Experience might be considered as a general law imposed upon nature by the cognitive subject. But what may be true of the general principle of causality is not true of the more particular laws of nature discovered by the natural sciences. This claim testifies to a confusion between nature in its formal sense (natura formaliter spectata) and nature in its material sense (natura materialiter spectata) and is accordingly a conflation of the idea of a pure science of nature, which investigates the a priori conditions of a physical world, and the empirical science of nature, which investigates the particular laws of nature (cf. Kant 1997: 47-65). Kant agrees that the general principle of causality is a necessary law of nature (in its formal sense) that can be known wholly a priori. Nevertheless, he flatly rejects the idea that the particular laws of nature, such as the 2nd law of thermodynamics or the orbit of Uranus could be equally known in an a priori way; on the contrary, their knowledge necessarily presupposes empirical research. Accordingly, while one might affirm that the general principles of nature in its formal sense are “imposed” upon it by the cognitive subject, this idea should be flatly rejected in the case of the empirical laws of nature. In the same way that man does not impose upon nature the existence of a mountain but rather discovers this existence empirically, man does not impose upon nature the 2nd law of thermodynamics or the orbit of Uranus, but rather discovers them through empirical research and theorizing. Thus, Bhaskar is wrong in assuming that his demonstration of the intransitivity of natural causation in the material sense goes against Kant’s ideas.

143

(4) The products of the activity of this ‘something’ are the events; this

something ‘generates’ the events.

(5) This ‘real something’ that is over and above events and independent of

them, and which, by its activity, generates the events, must be thought of as

a generative mechanism.

Thus, one arrives to one of Bhaskar’s most famous concepts: the concept of a

‘generative mechanism’. Bhaskar’s transcendental analysis has established the

existence of something real beyond events, whose activity and operation

generates the events. Bhaskar notes:

The world consists of mechanisms not events. Such mechanisms combine to

generate the flux of phenomena that constitute the actual states and happenings of

the world. They may be said to be real, though it is rarely that they are actually

manifest and rarer still that they are empirically identified by men. They are the

intransitive objects of scientific theory. They are quite independent of men – as

thinkers, causal agents and perceivers. They are not unknowable, although

knowledge of them depends upon a rare blending of intellectual, practico-technical

and perceptual skills. They are not artificial constructs. But neither are they Platonic

forms. For they can become manifest to men in experience. Thus we are not

imprisoned in caves, either of our own or of nature’s making. We are not doomed to

ignorance. But neither are we spontaneously free. This is the arduous task of

science: the production of the knowledge of those enduring and continually active

mechanisms of nature that produce the phenomena of the world. (Bhaskar [1975]

2008: 47)

144

It is the continuous and enduring operation of these normally invisible

mechanisms that produces the actual events of which we have experience. The

generative mechanisms of nature are in a state of continuous activity. This

activity, however, is not always realized. The reason for this is that the activity

of each mechanism is often impeded by the activity of other mechanisms. The

world can be seen as an arena in which generative mechanisms in constant

operation struggle to produce their natural effects. Thus, as Bhaskar notes, a

mechanism may be active but may not be realized or actualized. And it may be

actualized, but its actualization may pass unnoticed by men. Therefore, men

have empirical access to only a small portion of the total activity of the world’s

mechanisms. Nevertheless, what sciences seek to accomplish is to acquire

knowledge of these enduring mechanisms of nature. For this to happen, there

needs to be in place the transitive dimension of science in its social and

historical aspects.

Bhaskar insists on the fact that all objects of scientific inquiry have a

transitive dimension. As I have discussed, constant conjunctions, which are the

indications of the operation of a generative mechanism, can be normally

observed only in the context of experimental activity. And it is quite evident

that the setting up of an experimental framework presupposes a certain stage of

social and historical evolution.

But, even at a more basic level, the level of experience, there are

sociological and historical conditions for scientific inquiry. As Bhaskar notes,

“experiences and the facts they ground are social products” (ibid: 57). All

experience presupposes a theoretical-conceptual framework that mediates the

empirical apprehension of the world. There is no experience without concepts:

145

without concepts we can see nothing; according to Kant’s famous dictum:

“intuitions without concepts are blind” (Kant 1999:193-194). The conceptual

equipment of man is not given naturally but it is rather constructed through the

historical evolution of mankind and society.

3.8 From Generative Mechanisms to the Natural Tendencies of Things:

The Domain of the Real

Having established the intransitive existence of generative mechanisms of

nature and having demonstrated the necessity of the transitive dimension of

scientific research, Bhaskar then moves to the final step of his argument: the

analysis of how a generative mechanism of nature should be understood.

Bhaskar has purportedly proved that the world consists of generative

mechanisms, which exist independently of man. These mechanisms operate

continuously, even though they are not always realized and are rarely

experienced. Moreover, it is the operation of such mechanisms that produces

the actual events of reality, some of which we experience. When we set up an

experimental framework, what we seek is the isolation of a generative

mechanism so that we can observe what its operation is, when it is exercised

unimpeded by external interferences. Therefore, the statement of a causal law,

which is the ultimate aim of scientific theorizing, is nothing other but a

description of the operation of a generative mechanism.104

But what is a generative mechanism of nature? We know that is something

that is beyond the regular sequence of events that we can observe through

experimentation. Further, we know that it is something operative, efficacious, 104 “It is normally only in the laboratory that these enduring mechanisms of nature, whose operations are described in the statements of causal laws, become actually manifest and empirically accessible to men” (Bhaskar [1975] 2008: 46).

146

active, and that its operation produces events. At this point, Bhaskar makes a

crucial claim: this something must be primarily thought of as an agent, and

more particularly as a causal agent. Therefore, a generative mechanism must be

thought of in relation to the concept of a thing. Bhaskar here is asking us to

construct the notion of an active thing, viz. of a thing endowed with the power

to act. When we think of a generative mechanism, we should think of a thing

which acts: “For a generative mechanism is nothing other than a way of acting

of a thing” (Bhaskar [1975] 2008: 51).

It is at this point that Bhaskar moves to his neo-Aristotelian account of

philosophical ontology.105 A generative mechanism is the way of acting of a

thing. There are things endowed with powers. The things are naturally active;

they operate. More specifically, they operate in a continuous manner. They

continuously exercise the powers that they have. However, these powers are

only occasionally realized, because their operation is impeded by the operation

of other things. The way of acting of a thing, what a thing does, is determined

by the nature or essence of the thing. What science seeks to discover is the

nature of things, which by their continuous activity and interaction produce the

105 For the Aristotelian origins of his ontology of things (whose Aristotelian equivalent is the notion of a substance or ousia) and tendencies (whose Aristotelian equivalent is the notion of nature or physis), Bhaskar’s gives only a minor hint in footnote 34 (Bhaskar [1975] 2008: 50) of his book, where he refers to E. Anscombe’s and P. Geach “Three Philosophers” (1961), and more specifically to the chapter on Thomas Aquinas. Nevertheless, despite the undeniable affinity of his ontology with the Aristotelian ontology, Bhaskar almost never makes explicit reference to Aristotle’s thought. On Aristotle’s ontology of substance, see Irwin (1988: §§10-12; pp. 199-275); also Politis (2004: 190-259). On Thomist ontology, see Ch. 1 in “Metaphysics: A Theory of Things” of Stump (2003). The philosophical community has not focused on this relation between Bhaskarian and Aristotelian ontology, with the exception of the review by R.J.B. in the Review of Metaphysics (1980: 619-620), where it is claimed that Bhaskar’s “interpretation of the ontological conditions for science is thoroughly and unequivocally Aristotelian, although the author unexplainably avoids making this identification explicit. (Aristotle is mentioned only once, and then in a footnote). Bhaskar’s mode of argumentation is transcendental, but the resulting brand of realism is Aristotelian”. This relation must become the object of philosophical research in order for Bhaskar’s philosophy to be understood in its best light; Milja Kurki, who considers Aristotle’s theory of causation in her own works about causation in IR, has already worked to this goal in Kurki (2006), Kurki (2007), and, foremost, Kurki (2008).

147

actual phenomena of the world. As Bhaskar notes, “[S]cientists attempt to

discover the real essences of things a posteriori and to express their discoveries

in real definitions of the natural kinds” (Bhaskar [1975] 2008: 214).

Everything is endowed with certain powers determined by its nature, by

what kind of thing it is. According to Bhaskar these powers are to be

understood in continuous operation: they naturally exercise themselves. They

do not need to be triggered in order to get exercised. Bhaskar asks us to imagine

of a thing as a natural causal agent that acts naturally. It is for this reason that

Bhaskar asks us to think of a thing’s powers as ‘tendencies’; that is, as

potentialities which by nature tend to actualize themselves and which would

efficiently actualize themselves if they were not impeded by the contrary

operations of other causal agents. A thing by its nature operates, but only rarely

does its operation get actualized and even more rarely does its actualization gets

perceived by men.

Bhaskar can now describe what it is to state a causal law. The statement of a

causal law is, as has been argued, a description of the operation of a generative

mechanism. It is now clearer what it is to describe the operation of a generative

mechanism; it is to describe the tendencies of a thing, i.e. what a thing tends to

do intrinsically by its nature. A thing by its nature has its proper form of

operation, its proper activity. This operation is exercised even if it does not get

actualized.

Suppose that the nature of a thing determines it to exercise an activity that

we may call ψ’ing. Now, the thing will continually exercise this activity, that is,

it will continually be ψ’ing, even if this ψ’ing does not get actualized. If the

conditions are favourable it will actualize itself, and if the conditions are

148

favourable for us we can be able to perceive the ψ’ing. But in itself the activity

of ψ’ing is exercised continuously. Now, (a) if to state a causal law is to

describe a generative mechanism, and (b) if to describe a generative mechanism

is to describe the natural tendencies of a thing, then to state a causal law is to

describe the natural tendencies of a thing. Therefore, the statement of causal

laws implies the use of a distinct kind of conditional statements which describe

the exercise of the natural tendencies of things. Bhaskar calls this kind of

conditional statements ‘normic’ conditionals. Normic conditionals:

do not say what would happen, but what is happening in a perhaps unmanifest way.

Whereas a power’s statement says A would ψ in appropriate circumstances, a

normic statement says that A really is ψ’ing, whether or not its actual (or

perceivable) effects are counteracted. They are not counter-factuals, but

transfactuals; they take us to a level at which things are really going on irrespective

of the actual outcome. To invoke a causal law is to invoke a normic conditional.

(Bhaskar [1975] 2008: 51)

Bhaskar’s analysis leads to a further distinction concerning the levels of reality.

At the most immediate level, what we have is experiences. But, these

experiences can only be accounted for by reference to events in the actual

world. But now, one can see that events can only be accounted for by reference

to a more basic level: the level of generative mechanisms, of causal agents, of

things and of their natural tendencies to operate. Accordingly, one can divide

reality into three domains. The most basic level is the domain of the real. The

latter includes things which, by the specific nature that they have, tend to act in

certain ways in order to perform an activity that is proper to them. At the next

149

level, there is the domain of the actual, which is the level of events. The

domain of the actual is produced by the domain of the real, just as events are

produced by the tendencies of things, viz. by generative mechanisms. Things

naturally act in a continuous way, but their tendencies only get actualized in

relation to the activities of other things. The level of the actual is the level of the

realization of the tendencies of things, and it is only a subset of the level of the

real. Then, there is the least basic level: the domain of the experiences of the

events that take place in the actual domain. This is the domain of the empirical,

which is again a subset of the domain of the actual, and an even more limited

subset of the domain of the real.106

3.9 Conclusion

Through his use of transcendental argumentation, Bhaskar has arrived at the

point of providing a full-fledged ontology. He began with an inquiry on the

objects of scientific research. He claimed the existence of two dimensions of

scientific knowledge: the transitive one, which is man-made and socially

constructed; and the intransitive one, which is wholly independent of man.

Then, he tried to specify the nature of the intransitive dimension of science.

He asked how one should think the world might be like in order for our

cognitive activities (sense-perception and experimentation) to make sense.

Through this research he managed to provide us with an ontology in his sense

106 Here I adopt the following diagram which appears in Bhaskar’s ARST (Table 1.1, Bhaskar [1975] 2008: 56):

Domain of Real Domain of Actual Domain of Empirical Mechanisms √ Events √ √ Experiences √ √ √

150

of the term, i.e., through an investigation on the ontological presuppositions of

the intelligibility of scientific practice. Thus, he has claimed that the objects of

scientific research are neither experiences nor events nor regular sequences of

events. Contra Berkeley, he claimed that there are events beyond experiences.

Contra Hume, he claimed that the world has a structure beyond regular

sequences of events. Contra Kant, he claimed that this structure is a feature of

the world itself and not something imposed upon the world by the cognitive

subject. This structure of the world, which accounts for the regular sequences of

events that one observes through experimental activity, is to be understood in

the form of generative mechanisms producing phenomena. These generative

mechanisms are to be understood as the natural tendencies of things, which by

their nature tend to act and which are continuously exercised. Moreover, they

do so even if they are only actualized under appropriate circumstances and even

if they are perceived only at very limited circumstances. “The world consists of

things, not events” and science “is concerned essentially with what kinds of

things they are and with what they tend to do” (Bhaskar [1975] 2008: 51).

Things and their natural tendencies are “the real basis” of the regular sequences

of events that men co-produce in their experimental activity.

Bhaskar has moved from epistemology to philosophy of science and then to

ontology. The latter, however, has deep consequences in terms of the way one

should understand scientific practice. As a consequence, one returns to the point

of departure of the present chapter, which is the relation between philosophy of

science and scientific practice. Bhaskar’s philosophy claims to offer a different

image of the world than the one presented by the prevailing philosophies of

science. He alleges that his image of the world is richer, deeper and goes further

151

and beyond observable events and patterns. If the world is understood in a

different way, then science, whose aim is the understanding of the world, has to

be construed differently as well. According to Bhaskar, science has to search

far deeper than atomistic observable events; it should embark upon a quest of

the world’s operative but unobservable underlying structure, which constitutes

the causal generative mechanisms that produce the phenomena of the

observable world. Therefore, it is easily understandable that the adoption of

Bhaskar’s meta-theoretical principles would have an impact on the way

working scientists understand the subject-matter of their own field.

Overall, in this chapter, I presented, dissected and critiqued Bhaskar’s CR,

with special focus on its ontological assumptions. This was an endeavour that

had to be undertaken prior to the examination of the alleged usefulness of CR to

IR as a meta-theory on the grounds of which one might build ‘legitimate’ IR

theories. This was necessary, because what counts as SR in the context of the

meta-theoretical debates in the field of IR is mainly (albeit not exclusively)

Bhaskar’s CR. It is now on the subject of the relationship between CR, SR and

IR that I will turn my attention to.

152

4

SCIENTIFIC REALISM IN INTERNATIONAL RELATIONS

4.1 Introduction

This chapter examines how the concept of Scientific Realism (SR) and its

associated set of claims about theory, knowledge and reality have been

employed in the field of International Relations. As noted in the second chapter

of the thesis, SR is primarily a view about ontology. Consequently, it is

important to keep in mind that in this context the word ‘realism’ has its own

distinctive use and it should not be confused with the same word in political

realism and neorealism.

Attention to SR was drawn when Wendt (1987), in the context of offering

priority to ontology over epistemology, put forward the doctrines of SR as a

meta-theory for IR. Wendt wanted to develop a non-positivist social science of

IR; consequently, he used SR in order to provide the foundation for challenging

the positivist orthodoxy in IR. His ideas were most fully articulated in his

Social Theory of International Politics (Wendt 1999). Subsequently, by

bringing into the discussion well-worn debates from the philosophy of the

social sciences, other authors have supported or criticized his position.107

Intriguingly, Wendt’s discussion of SR has been judged to be “required

reading for any student of international relations, or political science for that

matter” (Krasner 2000: 131). Despite the fact that I would not like to flatly

dismiss this claim, I will argue that Wendt’s discussion puts forward SR as part

107 For example, Alker (2000), Jackson (2011), Krasner (2000), Patomäki & Wight (2000), Smith (2000), Wight (1999), and others.

153

of the articulation of a substantive meta-theoretical agenda. The cost of doing

this is that it distorts the genuine SR position and that it abuses SR’s status as a

philosophy of science whereas, in fact, SR does not represent unequivocally a

‘real scientist’s own philosophy’, nor do scientific realists themselves prescribe

that it should be taken as such. On the other hand, one may argue that such a

distortion and abuse may well be justified if Wendt’s ultimate aim is neither

objective exposition nor accuracy (after all, in his 1987 paper, Wendt mentions

absolutely none of the arguments against SR, not even the standard ones108), but

instead, a programmatic strategy of inspiration, so to speak, against both world-

system theory and neorealism. Such a line of argument would hold that Wendt

aims for a new meta-theoretical doctrine, which will motivate new ways of

studying IR. Indeed, Wendt’s emphasis on the ontological-realist features

which, in his view, an IR meta-theory should have if debates are to move

forward, has given further impetus to those perspectives in IR that privilege

ontological questions over epistemological ones (the latter tending to end up in

social constructivism). A focus on ontological matters is taken to point beyond

the alleged theoretical, methodological and praxiological cul-de-sacs in which

current IR theory finds itself. Choosing ontology does not imply the defence of

a specific ontological framework within which one should operate; on the

contrary, as Patomäki & Wight note,

It is not simply a scientific ontology we mean here, as in theoretical disagreements

over whether states are the most important actors, for example. What we mean by

108 See section 3.2 of the previous chapter.

154

ontology is a philosophical ontology; an inquiry into which is logically prior to the

development of any scientific or social ontology. (Patomäki & Wight 2000: 215) 109

The distinction between philosophical ontology and scientific ontology110 is

apparent in the following extract by Bhaskar:

Two senses of the term can be distinguished: either (a) the branch of philosophy

concerned with the nature of existence or being as such, apart from any particular

existent objects (philosophical ontology); or (b) the entities posited or presupposed

by some particular substantive scientific theory (scientific ontology). (Bhaskar

2003: 442)

Thus, one ought to take special note of the context in which the SR doctrines are

developed. Since Wendt brings SR into debates with their own dynamic and

logic, perhaps one should not expect that his promotion of SR should amount to

anything more than the offering of guidelines to navigate amid contradictory

meta-theoretical IR debates.

Consequently, one might claim that a full and just critique of Wendt et al.

should not restrict itself to the arguments and counter-arguments for SR, as

these presented in the second chapter, but should have a deeper understanding

of how specifically SR is meant to be taken up in the field of IR meta-theoretical

debates. It goes without saying, however, that such a perspective on SR’s

critical analysis runs the risk of turning SR into a series of self-conscious

examinations of IR as a distinct discipline as well as an object of study. Such

considerations would inevitably move one too far from the evaluation of SR 109 Patomäki & Wight point to Bunge (1996) for further elaboration of this point. 110 Two interesting papers on ontology are Michel (2009) and Jackson (2008b).

155

itself. As a consequence, some crucial caveats are in order. The development

of these caveats will be the focus of the next section of the chapter.

4.2 Some Theoretical Caveats

The discussion in this section of the chapter will put section 4.3 in perspective,

by explaining that the subject matter of the present chapter is not (nor, more

importantly, should it be) the evaluation or the reform of the IR meta-theoretical

framework in which the discussion about SR in IR takes place. Instead, I am

concerned with the evaluation of SR as this doctrine is found in this area of

study; in other words, my aim is to examine its credentials, its coherence and,

above all, its alleged indispensability in the IR meta-theoretical discussion.

First, it is important to understand in detail what exactly SR is opposed to in

the context of the IR meta-theoretical framework. Since, for example, Wendt

(1999) meta-theoretically rejects anti-realism, materialism, post-positivism,

empiricism and instrumentalism, it can be argued that these opposing

viewpoints deserve an adequate characterization (Chernoff 2002: 190).

However, for the purpose at hand (i.e. the evaluation of the introduction of SR

into IR), an elaborate and fully detailed survey will not be necessary. The

reason for that, as I am going to discuss in section 4.3 of the present chapter, is

that SR (especially the Bhaskarian version of it) is not introduced into IR as

merely one philosophical doctrine amongst other, more or less philosophical,

positions. Instead, it is introduced primarily as an instigator to action — SR is

meant to guide empirical research and motivate political action, not only to

elaborate philosophical argument. For instance, it is indicative of the role that

some critical realists envisaged for SR in IR theory that Kurki has recently

156

expressed her disappointment at a surfacing tendency for the depoliticisation,

fragmentation and de-concretisation of critical and philosophical IR research.

She argues that despite all its sophistication, the IR meta-theoretical and

philosophy of science debate has not offered any alternatives to either the

theoretical or the political/practical concerns of IR. This is an important failure,

not least because we are in the midst of a period which is characterized by a

multiple crisis of the national and international economic and political system.

As such, a “specialization in meta-theoretical inquiry has brought an interest in

a rather narrow and ‘technical’ set of philosophical questions at the expense of

broader political questions embedded in philosophical analysis” (Kurki 2011:

136).

Furthermore, one should not overlook what is taken to be SR’s most

significant feature when brought into IR: the fact that SR harbours its own

“enormous potential for those who wish to construct an account of the influence

of unobservable structures … on behavior” (Smith 1996: 36-7). In other words,

a secondary dimension of the defence of SR focuses on the key-word

‘unobservable’ —a first order theoretical matter— and (as I will show) this is

where the true rationale for introducing SR lies;111 that is, one does not merely

defend a philosophical doctrine or an attitude, but one also defends the

reification of specific first-order theoretical terms in IR theories (states, systems

etc.) whose employment and interpretation are dependent on meta-theoretical

choices and assumptions.

According to scientific realists, the decision on how to answer a meta-

theoretical question (“Should we use unobservables in our theories?”) should

111 Chernoff (2002), an otherwise excellent critical account of Wendt’s SR in IR, seems to have totally missed this dimension.

157

not be conflated, at least prima facie, with the first-order theoretical question of

whether specific suggested entities actually exist (and hence are fruitful or not).

This literally absolves one from carrying out a detailed examination of SR-

alternatives in IR (at least for the purposes of this chapter)112, since bringing SR

into IR also serves ontological questions (the ontological status and the identity

of specific unobservables) and not merely methodological agendas. Instead, one

should focus on the concept of ‘unobservable’ itself, and ask whether the use of

‘unobservables’ as a theoretical term legitimates (or even motivates) the

adopting of SR (section 4.3).

For reasons of completeness, nevertheless, I shall briefly examine the fact

that the two theories that SR opposes are social constructivism and

methodological positivism. The former encapsulates quite straightforward ideas

about how agents create the social world, how language determines social

ontology (reality) and epistemology (knowledge), and how a hermeneutical

stance accords better (or not?) with the rule-based nature of social/international

structures. The latter (opposed by both constructivism and realism) is a rather

complex meta-theoretical position, which combines realist and anti-realist

elements 113 with classical logical positivist ones. Positivists are usually

committed to objectivism, empiricism and naturalism; occasionally to

behaviourism, too. For my purposes, I will use the more or less accurate

description of methodological positivism proposed by Brglez.114

112 See Chernoff (2002) for part of the task that is side-stepped here; he argues that Wendt has misrepresented his opponents’ views. 113 Again, Patomäki & Wight (2000): 215-9 provides adequate grounds for the identification of positivism with anti-realism (a fact well-known in the philosophy of science, though apparently not in IR). 114 According to Brglez: “With a substantial amount of simplification (perhaps even unjustified about various positivistic methods, methodological strategies and research designs) methodological positivism inter alia assumes at the level (or the discourse) of the philosophy of (social) science an epistemological belief in foundations (objectivity), derived either from

158

Having said that, it is important to acknowledge in advance that whether in

choosing opponents or in accumulating defenders, SR has been more or less

received into IR under certain quite strong assumptions, which go unquestioned

in the relevant literature and which actually dominate the current evaluation of

SR doctrines. These assumptions make the reception of SR in IR meta-theory

seem easy, natural and perhaps even inevitable. First in the hierarchy of these

assumptions is the idea that the study of world politics and international

relations should strive for the status of a ‘science’ akin to natural science: IR

should be approached as a ‘scientific’ subject. I ought to mention these

assumptions, however briefly, since they constitute the basic framework that

facilitates the transfer of SR philosophy from the natural to the social sciences.

Furthermore, if any discussion of SR in IR in this chapter is going to be fruitful

for IR meta-theory itself, these assumptions must be put in the proper

perspective.

The end of the Cold War defined a new intellectual space for scholars to

challenge existing theories of international politics. Constructivists (e.g.

Nicholas Onuf, Friedrich Kratochwil, Alexander Wendt) drew on established

sociological theory in order to demonstrate how social science could help IR

scholars understand the importance of identity and norms in world politics.

Consequently, IR theorizing usually begins – although this was not the case in

the past – with the demand for a social theory or a road map that offers a

empirical experience (senses) or from human rationality (logic). Such an account of foundations at the level (discourse) of social theory leads epistemologically towards an inductive or deductive search for scientific laws (based ontologically on constant conjunction of events), and at the level of empirical analysis towards the correspondence theory of truth. At the same time, methodological positivism inter alia ontologically assumes an instrumental use of theoretical concepts and models (‘as if they were true’) at the levels of social and IR theory, and axiologically strictly separates ‘is’ from ‘ought’ in empirical research.” (Brglez 2001: 343)

159

scientific account of the social world, upon which rests the theoretical

understanding of international politics. A social theory should flesh out an

ontology and an epistemology in the form of a scientific account, i.e. it should

contribute to the study of IR the way ‘ontologies’ and ‘epistemologies’ are

supposed to do in the context of a ‘scientific’ discipline. Such theories are to be

used to analyse the origins, development and consequences of norms and

cultures and are to be presented in “the form of most general social scientific

propositions” (Holsti 1991:166).

In addition, by taking it for granted that a scientific study of IR is possible,

IR theorists do not fail to respond to methodological questions in their attempt

to elaborate the meta-theoretical foundations of their research:

These kinds of questions, which do not refer directly to specific empirical

explanatory problems in the way that substantive theories do, are stuff of meta-

theory and meta-theoretical concepts such as ‘agency-structure’, ‘micro-macro’

and ‘time-space’. … In the social sciences substantive theories aim to generate new

empirical information about the social world, whereas meta- or sensitizing theories

and concepts are concerned with general ontological and epistemological

understandings; meta-theories and meta-concepts are designed to equip us with a

general sense of the kinds of things that exist in the social world, and with ways of

thinking about the question of how we might ‘know’ that world. (Sibeon 2004: 13)

Hence, overall, a strong attachment to a rather simplistic and traditional notion

of science is assumed throughout the whole debate about science in IR. This

allows for peculiarly uncritical attempts to adapt philosophical notions from the

natural sciences to the meta-theoretical study of IR. Characteristically, in this

160

context SR serves as a platform for ontological-oriented theorizing by assuming

that SR’s conceptual strengths in the philosophy of the natural world are

transported intact into the philosophy of the social world.

At this point, a number of important questions arise: Should one doubt the

above assumptions? Should one beware of ‘scientism’ in mainstream IR? Are

the defenders of SR confusing the scientistic with the scientific? All too often,

philosophical models of science, which are based primarily on physics, have

dominated the study of social phenomena. Still, the positivist conception of

science has never been devoid of harsh critiques. The several Methodenstreiten

throughout the very long history of positivism as the guiding philosophy of

social sciences have always focused on the incompatibility of the very nature of

positivism with the nature of the social world.115 Dilemmas such as the

‘epistemic fallacy’, the problem of induction, the alleged irreducibly open

nature of the social world, the necessity of ceteris paribus clauses, value-

ladenness and the theory-ladenness of observation, inevitably and inherently

haunt any positivist study, including that of IR. Indeed, the contrast implicit in

this picture is the one that gave rise to the Explanation/Understanding divide116.

One could justifiably argue, however, that such epistemological divides are

quite alien to the concept of standard SR itself. If anything, it is the ontological

divide which is liable to raise doubts: namely, an SR interpretation of the

natural world does not extend necessarily to the social world itself; hence, the

standard scientific realist arguments nesting in the philosophy of science are not

necessarily suitable for the specific debates in the social sciences. Even if social

115 See among others: Dilthey (1988); von Wright (1971); Adorno et al. (1976); and Habermas (1978). 116 See Hollis & Smith (1990). An interesting critical discussion of the Explanation/ Understanding divide can be found in Wendt (1998).

161

studies should emulate some standard conception of ‘science’ and, additionally,

even if SR is transferred to the philosophical debates by resource to the

methodology of such a ‘science’, one can still not avoid the brute fact that the

referent of a scientific theory of the natural world would still be non-intelligent,

whereas the referent of a social theory (an agent) would be intelligent.117 Again,

at the very least, this ontological division calls for a radical re-conceptualization

of SR before implementing the interdisciplinary leap of bringing it into the

realm of the social sciences; the methodological troubles of social inquiry being

only tangential to the coherence of SR per se.

Wendt et al.’s starting point, though, seems to lie in the conviction that

scientific research done in the SR spirit is the best way to get to know the

world. Moreover, in so far as the conceptual system that Wendt himself builds

can provide a new, fertile pathway to further IR theoretical studies, then SR

itself would be legitimated as the appropriate meta-theory. The circularity and

the insularity that these assumptions impose on the discussion are quite

troubling.

Once again, doubts crop up at this point: should these features motivate one

to put on hold any close examination of the reception of SR in IR? One should

emphasize that our understanding of SR in IR is determined by the meta-

theoretical choices that have shaped the relevant debate. Therefore, one

legitimately wonders if one should examine one’s credentials before proceeding

to analyze the reception of SR itself.

However, it is worth stressing at this point that the aims of this chapter do

not allow either for the doubting of the specific meta-theoretical context of the

117 See Hollis (1992) for a more detailed elaboration of this point.

162

reception itself118 or for the doubting of its (characteristically old-fashioned)

conception of scientific inquiry. The conception of scientific inquiry that has

been adopted by IR is old-fashioned for a number of reasons. To name only

one, Kratochwil is right when he argues that:

Historians and philosophers of science in the last generation have also

fundamentally changed the conception of science. The emphasis has decisively

shifted from simple ‘tests’ and the discovery of universal laws, to a concept of

science as a practice. Instead of logic and the logical positivist speculation about

the nature of ‘true’ statements which were part of, for example, Popper’s ‘Third

World’, we have come to realize that knowledge production has an important

practical and historical dimension that needs to be reflected upon.” (Kratochwil

2006: 6)119

Furthermore, SR has been received as part of the post-positivist agenda.120 As

such, it would be inappropriate to cast doubts on the conceptual credentials of

the framework which considers international politics to be a subject for scientific

study, while examining the way IR has itself received the doctrines of SR. This

is a far broader matter, which goes beyond the discussion of SR which I

undertake here. Consequently, a critical examination of the conceptual 118 The wider context is, of course, the context of the so-called third debate in IR. It has been going on between positivism and post-positivism since the late 1980s and it is over whether positivist theories of IR are valid, and, if not, what can be an alternative. 119 Kratochwil points to the works of Fuller (1991b) and Knorr-Cetina (1981, 1999), which “all focus on the process of knowledge production instead of on ‘ontology’ or field-independent epistemological categories” (Kratochwil 2006: 6). One might add Collins (1992), Giere (1998), and Pickering (1995), along with my own reflections on the concept of science as practice (which encapsulates the dimension of the historicity of science) in the second chapter of the thesis. 120 More precisely, critical realists consider CR – the Bhaskarian version of SR – to be part of the post-positivist agenda, because they consider CR incompatible with positivism. On the contrary, Wendt’s version of SR belongs to the same agenda only in the sense that it constitutes a ‘via media’ between positivism and interpretivism by attempting to bridge these two prima facie incompatible methodologies.

163

framework which considers international politics to be a subject for scientific

study has to be added to those ‘external’ matters, which touch on the evaluation

of SR only tangentially (e.g. the Explanation/Understanding debate). This does

not mean, though, that, for instance, SR cannot play a crucial role in the ‘reasons

vs. causes’ debate. Obviously, choosing SR as the appropriate meta-theory for

IR strongly biases the methodology towards causal explanations. This aspect of

applying SR to IR will be referred to again in sections 4.3 and 4.4. Nevertheless,

the aforementioned issues should not be taken to be definitive for the evaluation

of SR in IR meta-theory.

In any case, the wide diversity of the authors who defend SR and Wendt’s

specific employment of SR’s doctrines, plus the rather enthusiastic welcome

that Wendt’s discussion of SR has received in IR, is a good reason to carefully

examine and analyze Wendt’s interpretation of SR in this chapter. As an

important offshoot of this assessment, I will go on to provide a far more

detailed critical analysis of Wendt’s recent turn to Quantum Theory in Chapter

5.

My arguments in this chapter will appear as follows: In the next section (4.3)

I present the SR debate in IR. In the first subsection of section 4.3, I deal with

Wendt’s understanding and distinctive use of SR in the context of IR; in the

second, I will examine how SR is deployed in the work of Colin Wight who,

inspired by Roy Bhaskar and his paradigm of Critical Realism, attempts to

reform IR discussions accordingly. In the third subsection, I intend to discuss

arguments for and against SR which have been put forward by other IR scholars

who, being familiar with Wendt’s and Wight’s works, have got involved in the

164

relevant debate by offering their own, often very differing, approaches to the

SR problematique.

In section 4.4 of this chapter I try to come to grips with two important issues,

which have either directly or indirectly (in the case of the second issue) been

raised by the SR debate in IR. Namely, in the first subsection of section 4.4, I

deal with the causes vs. reasons issue, particularly in the light of SR’s

contributions to the discussion of causation in IR, whereas in the second

subsection, I critically present what Fred Halliday used to call ‘meta-theoretical

hypochondria’. The chapter finishes with the presentation of my conclusions in

section 4.5.

4.3 The SR Debate in IR

Alexander Wendt and Colin Wight have been the main protagonists in the SR

debate in IR and I will devote a substantial part of my discussion to the critical

examination of their positions. It seems to me that Wendt should have

precedence over Wight since the former was the first to acquaint IR scholars

with the SR problematique; therefore, I begin with a discussion of his work.

4.3.1 Wendt on SR

In this section, I will illustrate how Wendt’s IR Theory has evolved with regards

to the study of SR within IR, from the appearance of his paper on “The agent-

structure debate in International Relations Theory” (1987) onwards. The goal of

my analysis is to highlight the main thread of Wendt’s successive reflections on

SR and other relevant philosophical issues that pertain to the four papers of

165

Wendt’s, which I intend to discuss here. 121 Notwithstanding the significance of

the philosophical issues per se, in what follows I will also explore how these

meta-theoretical assumptions underpin the Wendtian IR Theory. The reason for

doing this lies in the fact that one must always bear in mind that in Wendt’s

theorizing “meta-theoretical and theoretical matters are necessarily connected”

(Guzzini & Leander 2006a: xvi).

Wendt (1987) introduced SR into IR in his attempt to resolve the agent-

structure problem in IR Theory and thus come-up with a new IR Theory, which

could supersede the existing ones. He was motivated by his disappointment

with the inappropriateness of Waltz’s neorealism (Waltz 1979) and

Wallerstein’s world-system theory (Wallerstein 1974) to provide a

comprehensive structural explanation of the roles of the state and the

international system as mutually dependent entities. He argues that the

weakness of both theories is due to the individualist and structuralist ontologies

of neorealism and world-system theory respectively. Wendt summarizes the

problem as follows:

[…] neorealism and world-system theory share a common, underlying approach to

the agent-structure problem: they both attempt to make either agents or structures

into primitive units, which leaves each equally unable to explain the properties of

those units, and therefore to justify its theoretical and explanatory claims about

state action…Notwithstanding their apparent aspiration to be general theories of

121 One could also add Wendt (1992) to this list. However, his reflections on anarchy do not include any new ideas about SR and this is the reason that I have not included him. What is really striking is the context of the following statements at the end of this paper, at least when they are read in retrospect and one bears in mind the direction that Wendt’s theorizing has followed after its publication: “Neither positivism, nor scientific realism, nor post-structuralism tells us about the structure and dynamics of international life. Philosophies of science are not theories of international relations.” (Wendt 1992: 425)

166

international relations, the individualist and structuralist ontologies of neorealism

and world-system theory preclude the development of such theories. In contrast, a

structurationist approach to the agent-structure problem would permit us to develop

theoretical accounts of both state agents and system structures without engaging in

either ontological reductionism or reification” (Wendt 1987: 349)

In order to frame such a structurationist approach to the agent-structure problem,

Wendt was in need of a meta-theory, which could conceive of unobservable

entities in such a way that would enable us to see state agents and system

structures as mutually dependent and co-constitutive – that is, liable to mutual

reproduction or even change – according to the development of their

interrelationship over space and time. Wendt’s attraction to SR started with his

interest in the notion of the unobservable, as the latter presented itself in the

relation between agent and structure. According to Onuf, Wendt was intrigued

on how an unobserved ‘thing’ called structure had ‘observable effects’ (Onuf

2012: 25); even though, one might point out that system theorists also did this,

i.e. they used to attribute ‘observable effects’ to unobserved or unobservable

structures. Given the above, it seems that, while unobservables are not a problem

for “working scientists who take unobservables for granted”, they are a problem

for positivist philosophers (ibid). Wendt’s embracing of this idea led to the

“immediate if unplanned effect of unleashing an increasingly rancorous debate

over the discipline’s philosophical grounding” (ibid.).

Indeed, by drawing upon the literature of the philosophy of science and

especially – however, not exclusively – on Bhaskarian CR, Wendt’s SR

legitimates the ascription of ontological status to unobservable entities like

generative mechanisms. The conclusion to the third chapter of the thesis, which

167

refers to the notion of generative mechanisms, is relevant here. There I argued

that: “This structure of the world, which accounts for the regular sequences of

events that one observes through experimental activity, is to be understood in

the form of generative mechanisms producing phenomena. These generative

mechanisms are to be understood as the natural tendencies of things, which by

their nature tend to act and which are continuously exercised. Moreover, they

do so even if they are only actualized under appropriate circumstances and even

if they are perceived only in very limited circumstances. ‘The world consists of

things, not events’ and science ‘is concerned essentially with what kinds of

things they are and with what they tend to do’ (Bhaskar [1975] 2008: 51).

Things and their natural tendencies are ‘the real basis’ of the regular sequences

of events that men co-produce in their experimental activity.” (see page 150 of

this thesis).

Furthermore, Wendt claims that another crucial characteristic of SR is the

prioritization of ontology over epistemology. He argues that empiricists

subordinate ontology to epistemology, for “what exists is a function of what can

be known experientially” (Wendt 1987: 352), whereas scientific realists accept

abductive (or retroductive) inference as justified, “if the entity in question can

produce observable effects, or if its manipulation permits us to intervene in the

observable world” (ibid: 352). Following Wendt, SR assumes that the

unobservable entities, which are theoretical terms of mature scientific theories,

do in fact exist. This is because scientists themselves consider them to exist

and, as a consequence, organize their experiments in accordance with their

theories:

168

Thus, the fact that we can use theories about the (unobservable) internal structure

of atoms to build nuclear weapons which can destroy cities is a good reason for the

realist to believe that such structures exist, as we understand them today. (ibid:

352)

The above thesis is compatible with structuration theory, for the latter

legitimates the scientific consideration of unobservable and irreducible social

structures within its own framework. This holds because the structuration theory

“conceptualizes agents and structures as mutually constitutive yet ontologically

distinct entities” (ibid: 360). In other words, agents and structures are co-

determined in the sense that “social structures are the result of the intended and

unintended consequences of human action, just as those actions presuppose or

are mediated by an irreducible structural context” (ibid: 360). It is obvious that

SR underpins the application of structuration theory in IR, as this is

demonstrated by the fact that Wendt uses the notion of unobservables in order to

speak of structures.

I now come to the implications that the above statements have for IR

theorizing. Regardless of whether or not structuration theory provides a suitable

solution to the agent-structure problem, SR challenges the dichotomy between

the methodological orthodoxy of empiricism, which speaks in the name of

‘science’, and the ‘un-scientific’ approaches to IR offered by hermeneutics and

critical theory. Although not an IR Theory in itself, SR aspires to be a meta-

theory upon which one can build different kinds of IR Theories, in the

frameworks of which one searches not only for causes in the Humean sense, but

also for generative mechanisms. The existence of the latter can be inductively

169

inferred by the observation of the emerging phenomena/effects, which these

causal mechanisms produce.

To sum up, in his 1987 paper on agents and structures in IR, Wendt made

two basic claims. The first claim consists in that his structuration theory

provides a way to combine a focus on structures (ideas, norms, culture, etc.)

with a focus on agents. This explains how the international system can be

produced, reproduced and even changed due to the mutual reconstitution of its

agents and structures, which is not self-referential, but which allows agents to

undertake actions that can change the existing structures. As a consequence, an

explanation is now on offer as to why anarchy in the international system cannot

be properly explained only by its reduction to states’ behaviours and the

distribution of power between them (as Waltz allegedly argues). Moreover, the

same line of thought could offer an explanation as to why the structure of the

international system cannot be conceived by its reduction to the principles

which organize the world economy, namely global capitalism (as Wallerstein

argues). The second claim holds that SR prioritizes ontology over epistemology

and methodology and allows for the study of unobservables (structures), namely

of many social phenomena “which traditional empirical social scientists have

been forced to ignore” (Ringmar 1997: 271).

Wendt’s STIP (1999) is a follow-up to his 1987 paper and consists of two

parts. In the first part, titled “Social Theory”, in order to frame the meta-

theoretical background of his IR Theory, the author embarks upon the

philosophy of natural and social sciences. In the second part, titled

“International Politics”, Wendt unfolds his social theory of international

politics.

170

No matter how strange it may seem to the IR scholars who are not familiar

with the corpus of Wendt’s works, it is real politics that have inspired Wendt to

develop an IR Theory which is consistent in all its levels. In particular, he has

been triggered by his belief that neorealism - perhaps even more so than realism

- is flawed, in the sense that it leaves little room for changing the structures and

the functioning of international politics. He claims that one should take on

board the fact that the end of the Cold-War has proved that change is more

possible than neo-realists used to believe in the past (or even today).

Consequently, through his STIP, Wendt went on to frame a constructivist

approach to the study of international politics that provides for the possibility

for change of the morphology of the international political landscape.

Indeed, power politics is assumed to be the overwhelming practice in

international politics and realism is considered – especially by the majority of

mainstream American IR scholars – to be the theory that best describes that

practice. As a consequence, Wendt argues that one has to grasp “the link

between the ‘common sense’ in international affairs (mostly realist materialism)

and the way it not only understands but also constructs the actual state of world

politics” (Guzzini and Leander 2006a: xviii). In order to achieve this goal,

Wendt challenges realism. In order for his challenge to be successful, Wendt

puts the philosophical underpinnings of realism into the limelight. He grapples

with the meta-theoretical assumptions of IR Theory by grounding the building

of a substantive IR Theory on a scientific realist meta-theory. In doing so,

Wendt puts forward a theoretical fabric whose levels are intrinsically connected

in a way that “changing one can affect all the others” (Guzzini and Leander

2006a: xviii).

171

Wendt’s aims become even clearer if one pays greater attention to his

statement that the subject of STIP is “the ontology of international life” (Wendt

1999: 370). The understanding of the ontology of international life is a

prerequisite for the study of international politics, for “even the most

empirically minded students of international politics must ‘do’ ontology,

because in order to explain how the international system works they have to

make metaphysical assumptions about what it is made of and how it is

structured” (ibid: 370).

Wendt then goes on to propose a scientific realist meta-theory on which he

bases his new substantive theory of the international system. The latter is

differentiated from Waltz’s neorealism in that it views the international system

in idealist and holist terms instead of materialist and individualist ones. Wendt

defends an idealist or social ontology, so that he can attack the dominant

materialist ontology of mainstream IR theories such as, primarly, neorealism

and, secondarily, neoliberalism. This materialist ontology gives priority to

“material forces, defined as power and interest, and bring in ideas only to mop

up residual unexplained variance” (ibid: 371). However, it is important to note

that Wendt seems in fact not to endorse the ‘ideas all the way down’ thesis all

the way. Instead, he defends a thin constructivist approach, which is compatible

with the notion of a ‘rump materialism’, according to which the distribution of

capabilities (i.e. geographical and resource constraints) and the composition of

capabilities (i.e. technology) count as material phenomena (Wendt 1999: 109-

113). Therefore, Gofas (2007: 66) argues that “Wendt introduces the notion of

‘rump materialism’, in order to remind us that “constructivism should not

proceed ‘as if nature did not matter’ (Wendt 1997: 111).” Among the examples

172

of ‘rump materialism’ which Wendt provides is that of the German invasion in

Poland in 1939, which “was caused largely by ideas, but the material advantage

enjoyed by the Germans was an important factor about the situation: given their

aggressive intentions, it made it easier, and therefore more likely, for the

Germans to invade” (Wendt 2000: 166). All these are effectively summarized

by Wendt himself who writes that:

The key is to reclaim power and interest from materialism by showing how their

content and meaning are constituted by ideas and cultures. Having stripped power

and interest explanations of their implicit ideational content we see that relatively

little of international life is a function of material forces as such. It therefore makes

more sense to begin our theorizing about international politics with the distribution

of ideas, and especially culture, in the system, and then bring in material forces,

rather than the other way round. (Wendt 1999: 371)

Nevertheless, I am cautious over the claim that the relationship between the

material and the ideational forces is clear-cut in Wendt’s ‘rump materialism’. As

I have mentioned previously, Wendt does not seem to endorse completely the

‘ideas all the way down’ thesis. Indeed, one cannot be sure if he stands for ‘ideas

all the way down’ or for ‘ideas almost all the way down’, as Steve Smith (2000)

has shown by quoting many of Wendt’s contradictory statements on this issue.

Next, Wendt turns his attention to an exemplification of the ‘agent-structure

problem’. Namely, he examines the question of how the ideas held by state

agents relate to the ideas that make up the structure of the international system.

As it is known, methodological individualism, which is mainly expressed in

rational choice and game theoretical approaches, holds that “social structures

173

supervene on the properties of independently existing, preconstituted agents,

like states” (ibid: 371-72), but cannot see “that agents might be constituted by

social structures, that the nature of states might be bound up conceptually with

the structure of the states system” (ibid: 372). Therefore, Wendt proposes a

structurationist approach to the problem about how ideas and state agents are

interrelated. He argues that it is only through the application of the

structurationist approach that one can conclude that ideas ‘act’ upon states but

are also ‘acted upon’ by them. As Wendt puts it:

…the ontology of international life that I have advocated is “social” in the sense

that it is through ideas that states ultimately relate to one another, and

“constructionist” in the sense that these ideas help define who and what states are

(Wendt 1999: 372)

Wendt’s constructivism takes international system to be a social fact, which

allows for the possibility of social learning in the absence of which one cannot

conceive the reflexivity in the social world.

It is now time to deal with the meta-theoretical issues that underlie, underpin

and, to some extent, determine Wendt’s substantive IR theory. Traditional

constructivists used to claim that constructivist ontology is incompatible with

positivist epistemology (which is taken to be the mainstream epistemology in

natural sciences); instead, only interpretivist and post-positivist epistemologies

are considered to be compatible with constructivism. Wendt, on the contrary,

believes that even post-positivists are “tacit realists” (ibid: 373) in their

empirical research, since they test falsifiable conjectures of a kind similar to

those which are used by positivists in their own work; indeed, for him what

174

really matters is that “the epistemic authority of any scientific study, whether

using interpretive or positivist methods, depends on publicly available evidence

and the possibility that its conclusions might in some broad sense be falsified”

(ibid: 373). Therefore, Wendt suggests that ‘causal explanation’ should be

complemented with ‘constitutive explanation’. Such a development could

provide us with a ‘via media’122 as the appropriate way of theorizing about

causation in IR. The latter would, in its turn, make constructivist ontology

(based on a scientific-realist meta-theory) compatible with both positivist and

post-positivist epistemologies.

This brings up the discussion on questions of causation, which are addressed

by Wendt’s constitutive explanation. According to Wendt, “causal questions

inquire into antecedent conditions or mechanisms that generate independently

existing effects; this is generally what we want to know when we ask ‘why’

something happened or ‘how’ a process works” (ibid: 373). On the contrary,

“constitutive questions inquire into the conditions of possibility that make

something what it is or give the causal powers that it has, and as such they are

interested in relationships of conceptual, not natural necessity; this is what we

want to know when we ask ‘how is X possible?’ or, simply, ‘what is X?’” (ibid:

373).

Wendt, furthermore, is explicit in that natural and social scientists alike

should ask causal as well as constitutive questions. Nevertheless, the extent to

which a constitutive explanation could be useful for the study of political events

122 Gofas calls Wendt’s ‘via media’ an ‘epistemological Westphalia’, since by introducing this term Wendt attempts to transcend “current disciplinary divisions between causal and non-causal forms of theorising” (Gofas 2007: 67); the same term is also used by Wendt himself in a different context (2006a: 216).

175

of major importance to the historical development of the international system is

questionable. In order to address this issue, Wendt gives the following example:

Gorbachev’s New Thinking was a deep, conceptual reassessment of what the US-

Soviet relationship ‘was’. It was constitutive theorizing, at the lay level, and based

on it Soviets were able to end, unilaterally and almost overnight, a conflict that

seemed it had become set in stone. It may be that objective conditions were such

that the Soviets ‘had’ to change their ideas about the Cold War, but it does not

change the fact that in an important sense those ideas were the Cold War, and as

such changing them by definition changed the reality. (ibid: 375)

Admittedly, Wendt has at times attempted to explain the actual difference which

a scientific realist’s view of the social world makes for empirical research. The

above example testifies to his intention to do exactly that. Similarly, Ringmar

(2007) refers to Shapiro & Wendt (1992) where the two authors examine the

politics of consent from a scientific realist perspective, by taking on board the

existence of unobservable structures. Strict empiricists usually ignore the

existence of such unobservable structures and, hence, cannot come up with a

satisfactory explanation of consent. As a consequence, “empiricists can study

neither how political agendas are manipulated nor the processes through which

people’s preferences and identities are shaped” (Shapiro and Wendt 1992: 200-6,

quoted in Ringmar 2007: 274). In other words, due to the existence and the

impact of unobservable structures, agents cannot be taken at their own

estimation because they might be victims of false consciousness, something for

which empiricists cannot account for.

176

By analogy, I argue that sometimes a state does not behave according to the

distribution of power and wealth in the international system, because there are

one or more unobservable structures at work which impact upon this state’s

behaviour and thus co-determine it.

Following Shapiro and Wendt, Ringmar argues that in similar cases, one

needs “to make a hypothesis regarding what things would have been like if only

structural power had not been present, and then measure the difference between

this condition and the one presently at hand” (ibid: 274).

As already mentioned, constitutive theorizing, which Wendt encourages,

facilitates the reflective thinking on structures, which through their interactions

with agents make up various social facts. This opens up the possibility for

initiatives to be undertaken on behalf of the agents (individuals, states, NGOs

and radical political movements alike) in the direction of changing the

morphology of the current international political landscape. That given, the

dialogue between IR and the fields of Political Theory and Normative IR

should be intensified, since the perspective for changing the structures of the

international systems presents us with normative questions and practical

political challenges of the first order.

In this vein, Wendt claims that this dialogue between IR and the fields of

Political Theory and Normative IR has until recently “been kept very limited by

the domestic orientation of most Political Theory and the marginalization of

normative questions in IR by Realism” (Wendt 1999: 376-77).

This might hold with respect to the direction that IR studies have followed

mainly in the USA, but not in Europe. Furthermore, even in the USA we have

excellent examples of theoretical works where IR encounters classical and

177

modern Political Theory, Moral Philosophy and Ethics, Legal Theory and

Human Rights Theory.123 Furthermore, there are also recent publications where

FPA and EFPA are fertilized through their encounter with Political Theory and

Human Rights Theory. Suffice it to mention only Christopher Hill (2007), and

Lisbeth Aggestam and Christopher Hill (2008), where the authors, in order to

theorize about foreign policy-making in multicultural societies in general and in

EU in particular, first come to grips with questions about whether we have

different kinds of multicultural societies, what their specific characteristics are,

why and how have the latter emerged and, finally, how they do influence the

national foreign policy-making of the states within which such multicultural

societies have developed.

I would like to make one more point before completing my account of STIP:

since the possibility for the emergence of an ‘anarchical society’ is clearly

provided by Wendt’s theory, one may trace connections and affinities between

the latter’s ‘thin constructivism’ and the work of some theorists who belong to

the English School of IR theory. Wendt and most English School scholars seem

both to share a state-centric view of the international system and to assign to

ideas – although with differing degrees of emphasis - a role in the shaping of

world history. Furthermore, although from different perspectives the notion of

anarchy in the international system also interests them both - for instance,

Wendt (1992) was clearly a response to Bull ([1977] 2002). Suganami (2001)

argued that hypothetical causal narratives of the kind that Wendt produces may

assist historical research of the kind we encounter with in some works of the 123 An example of the attention paid to human rights within European IR Theory is Vincent (1986). Moreover, Michael Donelan ([1990]1992), Richard Tuck ([1999] 2001), Chris Brown (1992), (2002), (2010), John Rawls (2001), Kimberly Hutchings (1999) are only but a few of the works which belong to what one nowadays terms International Political Theory.

178

English School, while insisting that progress - beyond Wendt or the English

School - in historically and theoretically based IR scholarship cannot be

achieved “unless we engage in meta-historical reflections or critical

investigations into the nature and function of historical enquiry and

representation” (Suganami 2001: 419).

To sum up, the first part of STIP presents a constructivist meta-theory (based

mainly on the Bhaskarian CR),124 while the second half discusses notions such

as identity, norms and culture, along with those of anarchy in international

politics, which are closely related to various other streams of work in IR.

Moving on to the next major step in the evolution of Wendt’s manner of

doing IR Theory, I should like to reiterate that, according to Guzzini and

Leander, the update of Wendt’s IR Theory came “in the version of a theory of

history” (Guzzini & Leander 2006a: xvii), which can be found in his paper,

“Why a World State is Inevitable” (Wendt 2003). Guzzini and Leander argue

that Wendt in that paper favours a teleological approach to History, in which

politics and metaphysics are interrelated. On the one hand, I agree with them

that this work highlights a turning point in the development of Wendt’s IR

Theory, insofar as it explores the idea that collective entities, such as the world

state, might constitute kinds of self-organizing systems. On the other hand,

unlike Guzzini and Leander, I would not support the idea of a new “theory of

History”, but rather of a new ‘developmental theory of the international

system’. Nevertheless, in order to be fair to Guzzini and Leander, I must admit

that with the introduction of the notion of self-organizing systems, Wendt

124 Chris Brown and Kirsten Ainley speak of “a clear, albeit at times somewhat anodyne, version of constructivist epistemology” (Brown & Ainley 2007: 50). However, I don’t think Wendt’s meta-theory is so anodyne, since it has important implications for substantive IR Theory.

179

aspires to historicize his social theory of international politics (STIP); that is, he

aspires to put it into motion and thus further differentiate it from the through

and through scientific theory of neorealism.

It is interesting to note that it is Ringmar who first made this point, when,

after the publication of Wendt’s 1987 paper and before the publication of the

STIP, he argued that

Ηe [Wendt] tries, in a word, to put the neo-realist picture into motion, to historicize

it, to move it closer to actions and thought and to human life as we know it. Yet, in

the final analysis, Wendt is also a social scientist and not a historian, and while he

may try to historicize neo-realism, there are inevitably limits to any such enterprise

(Ringmar 2007: 285).

Given the above, one might say that Ringmar foresaw, to a certain extent, the

development of Wendt’s technique of IR theorizing towards a certain direction

that emphasizes notions such as personification, self-organizing systems,

collective consciousness, etc. (for instance, see Wendt 2003, 2004, 2006a).

In fact, Wendt takes on board the notion of emergence and argues that self-

organization theory provides a scientific basis for a teleological explanation of

the way the international system develops. The self-organization theory

challenges the neo-Darwinian theory of evolution, but only a few IR scholars

are today familiar with the former. Be that as it may, “in the social sciences

more generally the idea of self-organization has been around since the

‘spontaneous order’ tradition of the Scottish empiricists, and is getting

180

considerable attention today” (Wendt 2003: 492).125 Wendt has thus enriched

the concept of ‘constitutive explanation’, which has been coined in his 1987

paper and further developed in his STIP, with the notion of a spontaneous order

which may emerge in a self-organizing system “as a result of the interactions of

elements following purely local rules” (ibid: 498). By applying the above ideas

to IR Theory, Wendt concludes that the emergence of a world state is, at the

end of the day, inevitable. He holds it to be inevitable since, given the alleged

end-directedness of the historical development of the international system, a

global monopoly on the legitimate use of organized violence must also be

inevitable. However, it is important to point out that natural scientists, as well

as biologists who work on the ‘self-organization theory’, might reject the above

as not being a direct implication of the latter. Moreover, Wendt himself admits

that much of the work produced by self-organization theorists is not teleological

to the extent that “many self-organization theorists might vigorously reject any

such reading of their approach” (Wendt 2003: 492). In the context of his

approach, however, Wendt considers that agency is at both the micro- and

macro- levels absolutely consistent with his teleological explanation of the final

development of the international system to a world state:

125 For instance, Cornelius Castoriadis’ philosophical system is based on the notions of autonomy, emergence and self-organization. Castoriadis and the Chilean biologist, neuroscientist and philosopher Francisco Varela have been mutually influenced by each other’s work. They were both interested in the notions of autonomy and autopoiesis (which means self-creation, self-construction in Greek); see their discussion on this issue in chapter 4 (“Life and Creation”) of Cornelius Castoriadis’ book, Postscript on Insignificance – Dialogues with Cornelius Castoriadis (2011: 58-73), where the notion of emergence in physical sciences and biology is closely related to that of autopoiesis and autonomy in social and political theory. Moreover, Friedrich Kratochwil, in his critique of Wendt’s Social Theory of International Politics refers to Varela and his influence on Luhmann, when arguing that “constructivists can be found not only in cultural sciences. Indeed, some of its most prominent exponents have been biologists (Maturana and Varela 1992) who, in turn, influenced an entirely new system in sociology (Luhman 1984, 1997)” (Kratochwil 2006a: 22).

181

Human agency matters all the way down, but it is increasingly constrained and

enabled by the requirements of universal recognition. (Wendt 2003: 491)

The above brief presentation of the main ideas of his 2003 paper highlights a

number of issues. Firstly, it highlights that “it is through the process of collective

identity formation that Wendt sees the development of a world state as being

inevitable” (Brown & Ainley 2009: 76). Secondly, it shows Wendt’s attempts to

do justice to his SR assumptions, since ontology retains its priority over

epistemology and methodology. This is deduced, amongst others, from his

emphasis on the possible unobservability126 of the structures and boundary

conditions that are supposed to generate end-directedness in a system. However,

given that ontology comes first, one is faced with the hard epistemological

problem of “how we can know whether a world state is inevitable before the

system gets there?” (Wendt 2003: 503).

A crucial comment to be made here is that, according to Karl Popper, in

order for a statement to be so powerful as to claim the status of a scientific

conjecture, it must have a low probability of verification and the probability of

the statement being verified is always smaller the more specific the statement

is; and vice versa (Popper [1963] 1996). Therefore, I strongly doubt that a

statement as general as the one which holds that one day the international

system will inevitably develop into a world-state due to its alleged end-

directedness feature could have a claim to the status of a scientific

conjecture.127

126 Here unobservability is a great advantage, as no-one can refute Wendt’s assertions of telos/end-directness.

127In addition, one could not help underscore that in both his 2006a and 2010 papers Wendt appears to have abandoned the idea of the inevitability of the world state.

182

In 2004 Wendt published his paper “The State as Person in International

Theory” (Wendt 2004),128 which brought him closer to his latest proposal for a

‘quantum social science’ (Wendt 2006a, 2010). In his 2004 paper, which marks

his shift to the investigation of the notion of state’s ‘collective consciousness’,

he considers the state to be a kind of a ‘superorganism’ with its own ‘collective

consciousness’. As I discuss in chapter 5, it is exactly this search for a

comprehensive concept of ‘collective consciousness’, which has led Wendt to

develop his ‘collective quantum consciousness hypothesis’; the latter being one

of the cornerstones of Wendt’s ‘quantum social science’.

When it comes to the simple, albeit very important, question about how this

new way of theorizing about ‘the state as a person’ may be of some help to the

empirically-oriented IR scholars, Wendt is unable to provide any concrete

suggestions or empirical research proposals which differ from the existing ones.

He justifies his new theoretical work by invoking the rather trivial argument that

states should be persons for “states help bring order, and yes, even justice to the

world, and if we want to have states then it is better they take the form of

persons rather than something more amorphous, because this will help make

their effects more politically accountable” (Wendt 2004: 316).

To conclude, one should bear in mind that Wendt’s appeal to notions such as

‘self-organization theory’, ‘collective consciousness’ and lastly that of

‘collective quantum consciousness’ (the latter being part and parcel of his

‘quantum social science’ project) exemplifies how the meaning of the concept

128 For a critical examination of Wendt’s reflections on the state as a person, one can read the rest of the papers which co-constitute the “Forum on the state as a person”, which has been organized by the Review of International Studies: Jackson (2004a), Jackson (2004b), Neumann (2004) and Wight (2004).

183

of ‘constitutive explanation’ has been enriched and developed within Wendt’s

IR theorizing over the years.

4.3.2 Wight on SR

Colin Wight is a critical realist who is very interested in and very well exercised

about social science and philosophy of science issues. His Agents, Structures

and International Relations (2006) (henceforth ASIR) has been until recently –

that is, until when Patrick Thaddeus Jackson came up with The Conduct of

Inquiry in International Relations (2011) – the only book after Wendt’s STIP to

deal extensively with meta-theoretical issues. However, ASIR is not exhaustive

in the discussion of such issues. On the contrary, by considering the long

debated IR agent-structure issue as instrumentally useful for explaining the

development of the international system, Wight delves into the deep waters of

sociological theory in order to find a solution.

In the first chapter of ASIR Wight launches an attack against the ‘epistemic

fallacy’ in IR, which consists in the assumption that what is can be reduced to

what is known. Wight claims that positivism is mistakenly equated with science.

SR and the crucial issue concerning the affinities between natural and social

sciences, which allegedly legitimate the search for a common research

methodology, are also discussed in the framework of the first chapter. In the

next three chapters the author is concerned with the examination of the agent-

structure debate, first as this has unfolded in sociological theory, and then as it

has been introduced into and developed within IR. In the final three chapters

Wight focuses on the epistemological and methodological issues that are raised

184

by the consideration of the agent-structure problem. I will argue that Wight, like

Wendt, aspires to develop a scientific IR Theory.

To reach the aforementioned goal, Wight draws heavily upon Bhaskar’s

CR,129 whereas, on the contrary, Wendt’s SR is not as exclusively Bhaskarian.

Moreover, Wendt does not clearly designate himself as a critical realist. Talking

of Bhaskar, one may notice that Bhaskar is not very prevalent in the philosophy

of science literature. In the second chapter of this thesis, I have synopsized the

current debate about SR in the philosophy of science, without being, however,

able to trace any reference to Bhaskar. Monteiro and Ruby support the argument

that “Bhaskar does not feature prominently in the PoS [philosophy of science]

literature” and “indeed, the infrequent references to his work in the PoS are

highly sceptical of Critical Realism. See: Ruse (1981), Gunnell (1995), Kivinen

and Piiroinen (2004)” (Monteiro and Ruby 2009: 31, fn. 24).

In the same vein, Fred Chernoff argues that SR is not a form of CR but “CR

can be conceived of as a very specific, though quite unusual, sort of SR (perhaps

as Mormonism can be seen as a specific sort of monotheism)” (Chernoff 2007:

400) and he is right in further claiming that “the vast majority of scientific

realists in the philosophy of science would sharply disagree with the core

‘critical’ tenets of Bhaskar’s critical realism (as most monotheists would

disagree with the core of Mormonism)” (ibid).

Wight and Joseph argue that “CR is a very specific development of SR within

the human sciences” (Wight & Joseph 2010: 2) and that “it is vital to

differentiate SR from CR” (ibid: 4). Something which is really surprising,

129 However, more recently, Wight and Joseph have argued that “one could extend the notion of CR beyond the narrow confines of Bhaskar’s particular framework” (Wight & Joseph 2010: 2, emphasis added).

185

however, is that Wight and Joseph appear to ignore the complexities of the

ongoing SR debate within the philosophy of science, when they argue that

SR is a philosophy of science that arguably is the dominant approach now taken

within the philosophy of science. In fact the most serious contender to SR is Bass

Van Fraassen’s ‘constructive empiricism’, but this approach is fighting a rearguard

defence of an empiricist account of science; something that seems to most to be a

lost cause (ibid: 4)

Contrary to the above, Chris Brown uses SR and CR as synonyms:

Colin Wight’s attempt to distinguish between the two positions, associating

Bhaskar specifically with critical realism, is plausible but seems to me not to tally

with general usage, at least in the field of IR (Brown 2007: 409, fn. 1).

To conclude on this issue, I am aligned with Chernoff and make a distinction

between these two terms in my thesis, because although Bhaskar’s CR is in

conformity with several views and intuitions of other versions of SR, such as the

ones I have critically presented in chapter 2, it also departs from them in

numerous ways.

The next step in this discussion is to point out that Wight does not equate

science with positivism. He believes that the declared incommensurability of

differing epistemologies and the consequent lack of any ‘dialogue’ between

them is due to the fact that IR theorists - mainly but not exclusively of a

positivist vein - used to either prioritize epistemology over ontology or even

completely reduce ontology to epistemology. On the contrary, Wight gives

186

ontology a priority over epistemology (that is, we should care more about what

there is than how we know it), since he argues that it is in the historically and

socially driven ontological assumptions of the different IR paradigms where the

real differences between these paradigms lie. Thus, ontology is primary, whereas

epistemology and methodology are merely derivative, and not the other way

around. The inherent logic of this position is that since the differing ontologies

(for instance, those of Marxism and post-modernism) can be debated, the

possibility of a dialogue between the different IR paradigms should not be

excluded, meaning that the different IR paradigms are finally not

incommensurable and all equally subject to judgmental rationalism. Wight is

then an ontological realist (“there is a reality independent of the mind(s) that

would wish to come to know it” (Wight 2006: 26)), an epistemological relativist

(“scientific realism can be understood as an endorsement of Paul Feyerabend’s

methodological claim that ‘anything goes’” (ibid: 25)), and a methodological

pluralist, because he qualifies judgmental rationalism as the appropriate

methodology for his IR theorizing (“despite epistemological relativism, it is still

possible, in principle, to choose between competing theories” (ibid: 26)).

A critical point should be made here which concerns a subtle but decisive

difference between SR and CR with regard to a ‘correspondence theory of

truth’.130 It is noticeable that one finds in the philosophy of science clever

examples of why one must be skeptical about the SR’s endorsement of a notion

of truthfulness. It is true that even strict realists among the philosophers of

science, like Michael Redhead, identify shortcomings in the way SR justifies the

130 Suffice it to remind the reader at this point that “correspondence to reality – both in its observable and unobservable elements – is the scientific criterion for theory-choice. The goal of scientific theories is not only empirical reliability, but truth about (correspondence with) all aspects of the world” (Monteiro & Ruby 2009: 31).

187

notion of truthfulness. Redhead claims that, although he has nailed his colours

“to the mast as a realist”, still “there are a number of issues that still need

attending to” (Redhead 1995: 16). One of the four issues he quotes is as follows:

Realism seems to require some adequate notion of truthlikeness or verisimilitude.

Given that our theories are often discarded, can one nevertheless make sense of

approach to the truth? This is a very thorny technical problem in philosophy of

science that hinges on the question: What is a theory really about? For example,

consider an astronomical theory that predicts the number of planets P and the

number of days in the week D. Let one suppose that it gets both these numbers

wrong, but gets P + D right. Should this count in assessing whether the theory has

got closer to the truth? This is a question to which no totally satisfactory answer

has been given. Intuitively P + D is not an interesting or significant quantity to get

right, but how do we rule it out on purely logical grounds? (ibid: 19).

Having said that, one notes that the IR scientific realists who draw exclusively

on the Bhaskarian notion of SR (namely CR) 131 may be at odds with the idea of

a ‘correspondence theory of truth’. The reason is that CR distinguishes itself

from other versions of SR in that “it explicitly recognizes that the social nature

of knowledge means that science will sometimes get it wrong and that scholars

must therefore maintain a critical stance toward truth claims (Wight 2006: 39)”

(Monteiro & Ruby 2009: 31, fn. 24). The above conclusion does not come

without epistemological implications, for

131 See, for instance, Wight & Patomäki (2000), Patomäki (2002), Kurki (2006, 2008) and Wight (2006).

188

critical realism thus promises to combine relativism at the epistemological level

(making for pluralism by allowing all sorts of approaches, theories, paradigms,

research traditions, etc., to operate side-by-side within a discipline) and realism at

the ontological level (continuing to view scientific knowledge as getting at the way

the world really works, independent from our efforts to know it). (Monteiro &

Ruby 2009: 31, fn. 24)

In a critical mood, Käpylä and Mikkola argue that the attempt to combine the

fallibility of human knowledge with the ‘getting things right’ attitude based on

correspondence-like concepts of truth is problematic. Their insistence on this

position is evident in their conclusion where they write that “this is the case even

when Critical Realism endorses a more relaxed view of (what we called)

‘resemblance theory of truthlikeness’ that tries to overcome the problems of

correspondence theory” (Käpylä & Mikkola 2010: 32).

However, if those IR critical realists who object to a correspondence theory of

truth were aware of Worrall’s SSR, they would have probably been less reluctant

to accept any kind of theory that entails the notion of correspondence to reality.

Indeed, as I have claimed in the second chapter (page 109) of the thesis,

according to the moderate metaphysics of SSR, “there exists a structured reality

of which the mind is a part; and, far from imposing their own order on things,

our mental operations are simply governed by the fixed laws which describe the

workings of Nature” (Worrall 2007: 153). At the epistemological level the latter

implies that “not only is this structured reality partially accessible to human

discovery, it is reasonable to believe that the successful theories in mature

science – the unified theories that explain the phenomena without ad hoc

assumptions – have indeed latched on, in some no doubt partial and approximate

189

way, to that structured reality, that they are, if you like, approximately true (ibid:

154).”132 Due to the analysis above, one can clearly see to what misconceptions

about key-notions of SR the uncritical adoption of SR by IR may lead.

Next, there is a need for a second argument on Wight’s epistemological

relativism to be put forward, since Wight’s endorsement of Feyerabend’s

famous methodological phrase, “anything goes”, must be further explained:

It is only when epistemology swallows up everything, Wight explains, that that

phrase connotes a denial of judgmental rationality. Once, however, we separate

ontology from epistemology, the phrase takes on a different sense. It means that

although any individual research question may be judged rationally in terms of the

strongest argument, it is impossible to say in advance which methods apply.

Instead of methodological appropriateness being determined a priori by some

epistemological foundation, what makes a method appropriate or not is the

ontological nature of the object under study. The appropriate methods, therefore,

must be decided case by case. (Porpora 2007: 308)

I think this clarification with respect to the meaning of Wight’s epistemological

pluralism is of some significance, for other critics (cf. Kořan 2007: 325-26)

conceive that Wight’s position with regards to epistemology is that the latter is

simply reduced to empirical research. If this were the case, one need not have

any concerns about if and how epistemology could progress in the future.

However, the aforementioned is an ambiguous position to hold. When Wight

132 This critique also applies to Monteiro and Ruby who, in their attempt to show that SR cannot serve as the founding meta-theory of IR, seem to not be aware of SSR and all the relevant discussion I have presented in the second chapter. Thus they go on to argue that SR “hinges on accepting a highly contested claim: that science grants us access to a mind-independent world” (Monteiro & Ruby 2009: 32).

190

argues that “which structure does indeed dominate human practices at any given

time or place is an empirical question” (Wight 2006: 299) or when he writes that

“without taking some things as given, no research would ever get off ground”

(ibid: 249), he implies that it is ontology which defines methodology case by

case; not that epistemology does not or should not play any role in ontological

and methodological considerations. This is exactly the reason why Wight (2006:

249) quotes Hollis in arguing that

The proper conclusion is that epistemology has to go along way around, visiting

arguments about the historical particularity of all ways of searching into and

discovering truth but then returning with renewed determination to transcendental

questions of how knowledge is possible. (Hollis 1994: 259)

It must be admitted, though, that one cannot but comment that Hollis is so

general in his instructions on how epistemology could recover its role in the

future that the charge of absolute epistemological relativism in the case of

Wight’s epistemology has not been yet encountered in a satisfactory way.

Concerning the concept of ontology in Wight’s meta-theory, it is worth

mentioning that he is interested in two different kinds of ontology (which I

have already mentioned in the Introduction to this chapter): “philosophical

ontology, which concerns the relationship between the researcher and the

world, and scientific ontology, which concerns the catalogue of things that are

taken to exist and hence available to serve as objects of scholarly research

(Patomäki & Wight 2000)” (Jackson 2008a: 341).

One should also add that Wight’s ontology is an emergentist one in that it is

similar to that of Wendt (2003), since they are both Bhaskarian in their origin.

191

Ontological emergentism holds that at each layer of a stratified reality, properties

emerge which are unique and irreducible to the more fundamental entities of the

underlying layer.

From the aforementioned discussion of epistemology and ontology as these

are both meant within Wight’s meta-theoretical framework, one concludes that

Wight rejects Wendt’s ‘via media’, for the former is not interested in ‘bridging’

positivist and post-positivist methodologies. Moreover, Wight argues that the

explaining versus understanding debate (Hollis & Smith 1990) belongs to the

realm of methodology rather than that of epistemology. What is valuable for

Wight’s theory in the discussion above, is that the distinction between these

two terms is based on ontological considerations about the social world (Wight

2006: 256). Last, what I would like to repeat and emphasize further is that

Wight endeavours to reclaim scientific realism from the grip of positivism,

given that positivism in the eyes of even his opponents is equated with

science.133

All the above constitute a brief discussion of the meta-theoretical

considerations which constitute the bedrock of Wight’s IR theorizing about

structures and agents.

Now, I will discuss in more detail the substantive IR theory which Wight

unfolds in his ASIR. A difference of the outmost importance between Wendt

133 Similar comments have also been made by Douglas Porpora and Chris Brown, respectively: 1) “In IR as in other social sciences, Wight finds, science and positivism are so equated that challenges to positivism are construed as challenges to science. Thus, as we critical realists know, even positivism’s opponents think that in rejecting a positivist approach to human behaviour, they simultaneously are denying the purview of science” (Porpora 2007: 307), and 2) “One of the most important questions to be answered by any of the social sciences, including International Relations, is whether or not it is possible to study society using the methods of the natural sciences, suitably adapted. Those who believe it is are sometimes, and quite wrongly, referred to as positivists, even though they have little to do with August Comte or Ernst Mach; so pervasive has this usage become that that adherents of a scientific approach have partially adopted it, often describing their approach as neopositivist” (Brown 2007: 412).

192

(1999) and Wight (2006) is that the former used social theory as a platform for

exploring important questions such as the mutually constitutive effects of

interactions between states, the social structure of anarchy and the issue of

structural change. In contrast, Wight placed the agent-structure issue at the

center of his social theoretical project. Indeed, in order to study it, Wight brings

on board a great deal of sociological theory, ranging from Emile Durkheim,

Max Weber and Antony Giddens’ ‘structuration theory’ to Margaret Archer’s

‘morphogenetic approach’. However, contra Wendt, who relies on Giddens’

structuration theory in his attempt to tackle the aforementioned problem, Wight

separates CR from structuration theory and recovers “the analytical role of

social relations, especially those of some material nature” (ibid: 310). Thus,

Wight does not defend the ‘ideas all the way down' thesis but rather believes

that there is a strong irreducible material basis in some kinds of social facts or

phenomena.134

The employment of Wight’s emergentist ontology along with the

aforementioned social relational approach135 to the study of phenomena which

emerge at different levels (or layers) of the international system “allows for

incorporating the material dimension, inter-subjective actions, social relations

and pure subjectivity into a realist account of social ontology” (Kořan 2007:

324-25); in other words, according to Wight, there are the aforementioned “four

planes of activity”. Next, Wight distinguishes three types of agency:

134 This is probably due to the fact that Wight and other critical realists are ‘faithful’ to the Marxist origin of the Bhaskarian ontology. 135Moreover, it is very interesting that even in the Philosophy of Physics and Geometry, relationalism is recently enjoying renewed attention among philosophers and physicists who concern themselves with ontological questions about the relationship between space and matter. For instance, relationalists deny that space is ontologically prior to matter and seek to ground all claims about the structure of space in facts about actual and possible configurations of matter – see Belot (2011).

193

intrinsically subjective, inter-subjective and structure driven. Given the

aforementioned, it is rather a matter of the empirical research to judge which of

these three types of agency prevails (one notices that epistemology seems to be

reduced to empirical research, a point which is debatable as I have already

argued). Given also that agency is “embedded within, and dependent upon,

structural contexts” (Wight 1999: 109), but not reducible to structure, none of

these four levels of analysis is the most appropriate one for the study of the

agent-structure problem in IR in general. Consequently, since each time one

should choose the appropriate level of analysis in accordance to the specifics of

the case under examination, one cannot come up with a definite solution (one

that holds for all cases) of the agent-structure problem in IR.

Kořan reveals the following ‘flaw’ in the logic of the above ‘emergentist

solution’ to the agent-structure debate, which seriously challenges SR’s

assumption about the independent existence of the observed entities and has,

also, further serious epistemological implications:

By asserting that behaviour bears an imprint of structure while it is the agent alone

who decides to act, it indeed might follow that it is but an empirical question of the

relative autonomy of the agent. Yet, Wight himself is ambivalent on this issue. At

one point he insists that ‘social objects are as potentially impervious to the wishes

of individuals as are objects of the natural world (…) [h]ence it is misleading to

think that [its] importance depends solely upon whether they enter the minds of lay

actors’ (p. 271). On the other hand, Wight explicitly denies this view (p. 222) and

argues that the extent to which structures are actualized ‘will depend upon the

action, reaction and interaction of specific agents’ (p. 220). In both cases, whether

the structure dominates over agency, and whether it independently exists at all, is

194

not the researcher’s ‘empirical question’ but a matter of agential (in)abilities, and

thus solely a subjective factor. This poses a serious challenge to an attempt to

reconcile structuralism and agency under the realist roof, unless we treat the

‘structural’ emergent layer merely as an epistemological/heuristic device. (Kořan

2007: 325)

On the other hand, Brown and Ainley welcome the fact that, thanks to Wight,

one can avoid getting lost in the realm of abstract theorizing about the agent-

structure problem and instead use only empirical analysis in order to establish

the causal significance of particular agents and particular structures in

particular situations:

This is anathema to theorists such as Waltz, who want to simplify the study of IR,

but does seem to be the most reasonable way to think about agents and structures,

and, very importantly, it pushes us back towards real-world problems, and away

from those more abstract questions that are best dealt with by philosophers of

science rather than scholars of International Relations (Brown & Ainley 2009: 70)

I rather agree with Brown and Ainley that the solution to the agent–structure

problem which Wight suggests brings one back to real-world politics (and this is

really important since IR’s subject matter is primarily real-world politics, and

not sociological theory or philosophy). However, what is highly problematic is

not only the uncritical introduction of abstract theoretical terms belonging to

other disciplines into IR, but, first and foremost, the fact that IR seems to follow

a trend that has also been torturing other social and political sciences. This trend

consists in the extensive concern with the philosophical foundational problems

195

of these sciences, at the expense of the development of substantive theories. For

instance, Nicos Mouzelis, who criticizes the existence of this same trend in

sociological theory in his book, Back to Sociological Theory (1991), devotes a

whole chapter (“Philosophy or Sociological Theory?”) to its discussion and

argues, among others, that

[…], it remains possible as well as desirable to distinguish between a

predominantly philosophical and a predominantly theoretical-sociological

analysis…Even if sociological theory must always entail philosophical

presuppositions, it does have its own autonomy and logic. Conflating and thus

abolishing the distinction between epistemological/ontological and sociological-

theoretical issues proper therefore destroys the latter’s specificity and relative

autonomy – resulting in analyses which, despite the numerous insights and brilliant

comments they provide on particular theories, often lead to misleading or

inadequate conclusions. (Mouzelis 1991: 12-13)

The above should also hold for IR in the sense that, if one substitutes IR for

sociological theory, then one gets a concrete depiction of the state of IR after it

has been invaded by meta-theoretical concerns.

Furthermore, in this abstract Mouzelis replies to those who, like Wight and

Joseph, argue that one of the many good reasons that the social sciences must

engage in philosophical reflection is that they need to assert their scientific

status, because they are not so developed as the natural sciences, which “are

often said to have only fully matured when they left philosophy behind and

196

established themselves as autonomous modes of knowledge generation” (Wight

& Joseph 2010: 5).136

Taking a step further, I would like to argue that, by taking literally Wight’s

suggestion that one should take refuge in empirical research if one wants to see

how the relationship between agent and structure is shaped in each different

case, one risks concluding that there is no need to theorize about agents and

structures at all. My point, however, is that although Wight and Wendt have

heavily drawn upon both the philosophy of science and social theory in order to

build their own IR theories, they have put more emphasis on the former than on

the latter. Therefore, Mouzelis may be right when he argues that the constituent

elements of each of the aforementioned “four planes of activity”, which are

suggested by Wight, are very poorly theorized because

Wight’s emphasis on ontology and his lack of emphasis of the theoretical tradition

that Parsons initiated in the post-war period – a tradition which theorizes in a

rigorous and logically coherent manner such basic concepts as values, norms,

institutions, etc… – has led him to an ad hoc type of theorizing. (Mouzelis 2008:

226, fn. 3)

Therefore, it is my opinion that those IR scholars that are interested in the agent-

structure debate should continue theorizing about it by drawing upon

sociological theory. One could, for instance, take on board Mouzelis (2008), and

136 The assertion that natural scientists believe that they do not need philosophy of science in order to have their work done is not very accurate, since nowadays one observes an increased interest of physicists in the philosophy of science and especially in the ontological problems which are posed my modern physical theories, such as QM, relativity, quantum field theory and string theory. Thus, one could argue that physicists have become more interested than they have ever been before in the works produced by philosophers of science on the foundations of scientific theories and vice versa (see, among others, Healey 2009, Callender & Huggett 2001, Kuhlmann et al. 2002, Dieks 2006, 2008, etc.).

197

his most recent reflections on the agent-structure issue.137 Mouzelis’ work

reveals that apart from Parsons, Giddens, Bhaskar, Habermas and Archer, there

are also important – albeit not so well known beyond the borders of their own

science – sociologists, like Alexander and Lockwood, whose works should be

carefully considered in the context of the study of the agent-structure issue. In

the second chapter of the thesis, I have gone for a rather detailed discussion of

SR as it has been deployed in the philosophy of science literature. I considered

that to be a necessary task given the eclectic and uncritical use of SR in IR. I

could have done exactly the same by providing an analytical discussion of the

agent-structure debate within sociological theory in a separate chapter, however

such an endeavour would have been a distraction from my work, since the

subject matter of this thesis is SR and not the agent-structure debate per se.138

However, one should bear in mind that the discussion of the agent-structure

issue in sociological theory is far richer and more sophisticated than it has been

presented within Wendt’s and Wight’s works. Moreover, my emphasis on

Anglo-Saxon sociological theory is not accidental; the purpose of thinkers who

belong to this tradition has been to clarify conceptual tools and to construct new

ones by following criteria of utility rather than truth (this may also be useful to a

pragmatist turn in the study of the agent-structure debate within IR):

137 See, in particular, “Part V: Towards a non-essentialist holism” of Mouzelis (2008). 138 For a very clearly written overview of the agent-structure debate in IR (which is accompanied by a good bibliography to start with), see chapter 4 of Brown & Ainley (2009). To this I would add Ainley (2006), an excellent piece of work where the author is not caught up in perplexing meta-theoretical debates but, on the contrary, relates her work in international political theory to real problems in international political practice and, by emphasizing agency, explores alternative views of it. Thus, she comes up with outlining “models of agency as sociality and responsibility as a social practice, arguing that these models both better describe the way we talk about and experience our social lives, and also often offer significant possibilities to broaden the scope of international justice and enable human flourishing” (Ainley 2006: 3).

198

Viewing pragmatism in the manner of C. S. Pierce – i.e. as a method for the

clarification of concepts by showing how they are or can be used – one might argue

that sociological theory, as developed by Parsons and others, has as its major task

to clarify current conceptual tools and to construct new ones by following criteria

of utility rather than truth. Adhering to this orientation, I have kept clear of the

type of ambitious theorizing that purposes to provide substantive universal

propositions either in the form of ‘laws’ and contextless generalizations, or in the

form of philosophical analysis on the ontological nature of the social, the

possibility of social knowledge, the constitution of the subject, and so on.

(Mouzelis 1995: 8-9).

This way of theorizing leads to the development of middle-range theories, as

opposed to grand theories of the type Waltz, Wallerstein and Wendt provide. It

indicates a possible re-orientation of IR theorizing; this may happen if one

seriously considers that middle-range theories come up with certain research

proposals for empirical test/falsification of the utility of their conceptual tools, in

contradistinction to the fact that neither Wendt nor Wight have yet generated any

empirical (note: not empiricist) research programme in the field (cf. Griffiths,

Roach & Solomon ([1999] 2009: 156-57)).

Going back to the discussion of agent and agency in ASIR, I would like to

draw attention to how Wight addresses the question about whether a state can be

conceptualized as a person in its own right. As Porpora claims, given that “we

know from Davidson that the mark of agency is intentionality and from

Anscombe that intentional action can be understood as behaviour done for

reasons” (ibid: 311), one could further argue that states seem to behave for

reasons and should, therefore, be personified. However, Wight – and I

199

sympathize with his view – disagrees with the personification of the state by

counter-arguing that reasons and intentions imply a background of mental states

and this, in its turn, implies that the state should be taken to be a ‘superorganism’

with ‘collective consciousness’. This is a conceptualization with which he

cannot come to terms. On the contrary, Wendt has put on the table and defended,

first, the idea of ‘superorganism’ with ‘collective consciousness’ (in order to

argue for the allegedly inevitable development of such a ‘superorganism’ as the

international system to a world state – see Wendt 2003) and, second, the idea of

the personification of the state (Wendt 2004). Wight takes issue with Wendt’s

concept of ‘collective consciousness’ and proposes a different one, that of

‘collective intentions’. He does not, however, elaborate further on this concept

(cf. Porpora 2007: 312). In Wight’s own words:

Wendt’s adherence to ‘state-as-agent’ thesis, however, is an endorsement of

structuralism at the level of the state. As Bhaskar puts it, ‘nothing happens in

society save in virtue of something human beings do or have done. What this

means in relation to theories of the state is captured nicely by Bob Jessop, who

argues, ‘[i]t is not the state which acts: it is always sets of politicians and state

officials located in specific parts of the state system’.139 In Wendt’s theory states

not only take the place of persons, but actually are persons. That is, states play the

role of human beings for Wendt. But states, even if they are agents, are not

persons. On, the contrary, states are institutional structures constructed by human

beings. (Wight 2006: 187-88)

139 See Jessop (1990: 367).

200

To put it in a nutshell, Wight argues that the state should be seen as an ensemble

of structures rather than an individual agent. Brown and Ainley summarize his

argument along the following lines and make the additional point that FPA

should have much sympathy with Wight’s conceptualization of the state:

He [Wight] emphasizes that the way that state power is actualized depends on the

actions of agents – individuals and groups – located in groups within the ensemble

of the state (Wight 2004: 279). So, for Wight, the state is not an agent in itself, and

certainly not a person, but rather a body that facilitates the exercise of power by the

agents within it, a position with which foreign policy analysis (FPA), … , should

have much sympathy. (Brown & Ainley 2009: 76)140

Making three further comments will complete my discussion of ASIR:

First, Wight neither provides us with certain examples of how his various

accounts of structure and agency could lead to certain empirical research

questions, nor tells us what kinds of normative or analytical conclusions we

could arrive at through the realization of these accounts. Bjola is right when he

wonders about “what kind of norms, institutions, or social identities follow

from the conceptualization of structure or agency along the lines suggested by

Wight” (Bjola 2007: 317) and poses the following plausible questions:

Does an account of structure as ‘law-like regularities’ entail a configuration of

interests and identities that is analytically or normatively different from a view of

structure as ‘relations of difference’? Similarly, does a conceptualization of agency

as ‘positioned-practice places’ favor different outcomes than a conceptualization of

140 Indeed, though FPA should avoid disaggregating so much that the state loses all definition and the assumption of agency qua state.

201

agency as ‘freedom of subjectivity’? If the answer is ‘yes’ in both cases, then the

key questions (how and why) are left unanswered by Wight. (ibid)

In the same vein, Lawson in his book review of ASIR argues that “if the better

probing of empirical puzzles is a core dimension of Wight’s agenda, it is a

shame that he has contributed relatively little to them here” (Lawson 2007: 774-

75) and expresses his fear that “Wight’s extensive underlabouring” may

“become a means for meta-theoretical bunkerists to spend even less time on the

empirical front-line” (ibid).

Second, it sounds rather odd that Wight, towards the end of ASIR, claims

that he does “not intend to outline a general theory of international relations”

(Wight 2006: 294), although this is exactly what he appears to have done

throughout the whole book (Floyd 2007: 188). The logic behind this argument

is rather simple: given that in the Bhaskarian version of SR it is ontology which

determines both epistemology and methodology, Wight’s commitment to

ontological realism suggests that he aspires to build a general theory of

international relations upon it.

Third, Wight assigns to science an emancipatory role and a practical task;

that is, Wight aspires to provide a scientific IR that will help us understand and

analyze the fabric of the international system properly. Such an endeavour

might make this system’s shortcomings surface and thus encourage and

facilitate action in order to improve the way the international system works

nowadays. However, what differentiates Wight from Wendt is the former’s

strong emphasis on his argument that philosophical choices are finally

political. Onuf is clear with regards to this feature of Wight’s ASIP:

202

Yet Wight’s conviction that philosophical choices are finally political gives the

book a polemical feel; it was soon followed by an exchange between Wight and

Kratochwil that left some readers wondering if their differences warranted so much

posturing.141 (Onuf 2012: 25)

I will conclude my discussion of Wight by drawing attention to the following

three features of CR: first, the goal of science is to describe how things really

are, and, second, there is a world out there, which is independent of our minds

and our ability or inability to know it. Third, this world is inhabited by both

observable and unobservable entities, which are the objects of science. The

existence of unobservable entities is inferred from their observable effects;

therefore, unobservable entities are described with reference to their effects.

Finally, causation is real, and not simply the product of human thought (on the

contrary, according to any traditional empiricist, we are free to talk about the

existence of a ‘lawlike regularity’ but not about the existence of a ‘cause’) (cf.

Brown & Ainley 2009: 58).

4.3.3 Critique – Opponents of SR

Turning to the critics of SR, Buzan believes that “in line with the current

fashion in IR Theory, it is the philosophical side of Wendt’s theory that has

141 I suppose that Onuf refers to an exchange Kratochwil and Wight had had about epistemological issues in 2007; Kratochwil (2007a) had kicked off with his plea for “a pragmatic approach to theory building”, Wight (2007a) replied by arguing that “inside the epistemological cave all bets are off”, Kratochwil (2007b) came back with a paper having the telling title, “Of communities, gangs, historicity and the problem of Santa: replies to my critics” (his other critics have been Lebow and Suganami) and Wight (2007b) closed this rather “vicious circle” – besides the sophisticated arguments which both sides have put on the table - with the equally telling title, “A response to Friedrich Kratochwil: why shooting the messenger does not make the bad news go away!”.

203

attracted the most comment” (Buzan 2006, xv). This explains the fact that the

majority of the attacks which have been launched on STIP focus on SR, namely

the meta-theoretical bedrock Wendt has built his substantive IR Theory upon.

In other words, there are two distinct trends as to the critical examination of

Wendt’s STIP. On the one hand, those IR scholars who are genuinely

committed to the study of SR usually “believe that progress in IR Theory can

only be achieved by getting the whole enterprise onto a firmer foundation in

philosophy of knowledge” (Buzan 2006, xvi). On the other hand, those who

consider the debate on the philosophical foundation of IR to be not so decisive

for overall progress in IR Theory, focus more on the theoretical rather than the

philosophical aspects of Wendt’s STIP. I will focus on the critics of SR/CR as

an IR meta-theory, without, nevertheless, neglecting to track the implications of

the possible strengths and weaknesses that SR might entail for the development

of substantive IR theories. Namely, I shall be examining some of the main IR

critics of SR/CR such as Suganami, Kratochwil, Jackson, Chernoff and Lebow,

with a view as to how they see SR/CR as both an IR meta-theory and a basis for

first-order IR Theory building.

Friedrich Kratochwil, a leading constructivist with a very wide and solid

philosophical background, attacks Wendt’s SR on many fronts. The most

crucial issue he raises is whether or not Wendt’s version of realism is

compatible with constructivism as a meta-theoretical orientation. According to

Kratochwil, Wendt seems to neglect the social component of SR, which

consists in science being conceived “as a communal practice rather than as a

simple finding and ‘testing’ of theories” (Kratochwil 2000: 36). This is even

more surprising since Bhaskar, upon whose work Wendt relies, sees talking and

204

debating scientific issues to be a part and parcel of the process through which

scientific progress is achieved.

Furthermore, as discussed elsewhere in the thesis, the three basic premises of

SR are the following:

1) the world is independent of the mind and language of individual

observers;

2) mature scientific theories typically refer to this world;

3) even when it is not directly observable. (Kratochwil 2000: 37)

Kratochwil sees no problem in considering “unobservables” to be (natural or

social) entities, since “that there are ‘unobservables’ in every theory is hardly

controversial, strict empiricists excepted” (ibid: 38). Moreover, Kratochwil

holds, that “hardly anyone – even among the most ardent constructivists or

pragmatists – doubts that the ‘world’ exists ‘independent’ from our minds” (ibid:

38). The problem is how we can come to know this independently existing word.

According to Kratochwil, one should take on board that this world cannot be

recognized and conceived in its totality in a pure and direct fashion, i.e. without

any ‘description’. We do not know, he goes on, whether “what we recognize is

already organized and formed by certain categorical and theoretical elements”

(ibid: 38, italics are mine). This leads us directly to the Kantian notion of “thing

in itself”. A “thing in itself” remains unrecognizable and unknowledgeable

unless it is brought under a description, something which brings us to the

following two significant questions:

205

First, whether this ‘naming’ is indeed a function of the congruence of our concepts

and the ‘things’, as this provides the yardstick for ‘truth’; or, second, whether

‘truth’ is a matter of the conditions governing the justifiability of assertions rather

than a matter of the ‘world’ (ibid: 38)

In his attempt to address the above two questions, Kratochwil argues that “things

in themselves” are knowledgeable only in the context of the social practices that

make the things into what they are called or referred to; consequently, ‘truth’ is

not a matter of the world. Given that Kratochwil questions “the existence of

‘truth’, along with ‘science’ as method of arriving at it” (ibid: 43), one comes to

conclude that the search for a foundational meta-theory in IR is infeasible and

meaningless. As a further consequence, SR is also incompatible with the kind of

constructivism Kratochwil defends. Next, in order to avoid the accusation of

relativism or even nihilism (although such an inference would be unjustifiable,

given that the aforementioned critique concerns and applies only to a very

dogmatic epistemology and concept of science), Kratochwil takes refuge in a

wider conceptualization of science, in which the use of the notions and rules of

logic (bivalence or rational principles, deduction, etc.) are not sufficient to help

one grasp the reality in all its aspects. After all, Kratochwil argues, the “finality

and legitimacy of a judgment coincide only in logic” (ibid: 44). In his attempt to

provide a solution to this deadlock, he brings ethical principles into play that

interpret the question of decidability as one of a fair procedure, which provides

an honest approach to scientific debates as well as an honest judgment about the

existence or not of scientific progress. I think this is a remarkable point

inasmuch as Kratochwil reminds us that even the late Popper moved towards

accepting that purely logical criteria are not sufficient to guarantee truth and

206

scientific progress. Instead, a conduct of rational discussion among scientists

that should be based on certain ethical rules must be introduced.142

This is exactly the social component of the scientific progress, which I have

mentioned at the beginning of my presentation of Kratochwil’s perspective of

science; it is in the lack of this feature that Kratochwil identifies one of SR’s

major shortcomings.143

Kratochwil’s piece shows clearly how differently – sometimes even

idiosyncratically – Wendt’s SR is conceived by his critics. Two examples

should suffice to make this point more eloquently:

First, in Wendt’s seminal (1987) paper, where SR is introduced in the study

of international relations for the first time as the proper meta-theory for the

concerns of IR, the examination of the arguments against SR is rudimentary, if

not entirely absent. This makes for a lop-sided presentation that obscures both

the strong claims SR makes, as well as the process in which its conceptual

resources can be employed to defend SR from anti-realist attacks. Perhaps this

poor portrayal of SR explains why a lot of Wendt’s critics and commentators

(e.g. Suganami; see my second example) find SR unnecessary, even though

they accept most of the traditional SR claims.

The second example I have in mind is where Suganami (2002) criticizes

Wendt, based on nothing more than personal intuitions and opinions, instead of

an actual critical analysis. This leads to a variety of rather simplistic claims. For

142 Look at Kratochwil (2000: 39). A similar point to Kratochwil’s, however not identical, is made by Lebow when he argues that “Popper,…, ultimately came to understand science as an ethical practice and rejected altogether the idea that philosophy could provide truth warrants” (Lebow 2011: 1219). 143 For a reply to the Kratochwil’s critique of Wendt, from the perspective of CR, see Morgan (2002). Morgan argues that Wendt’s defence of SR “is relatively untouched by Kratochwil’s critique, since that critique is constructed in terms of empiricist conflations” (Morgan 2002: 118).

207

instance, Suganami claims that “Of course, all our scientific assertions can be

construed realistically, that is, as involving ontological commitment on the part

of those who make such assertions” (Suganami 2002: 27, italics are mine), but

this cavalier attitude toward SR misses the fact that the realism debate is exactly

about how to cash out assertions such as the one above; that is, explaining what

‘realistically’ in fact means. In the same vein, Suganami writes that “scientific

assertions can be said to depict the world as it is independently of our

knowledge claims only if they are not wrong, and we can never be sure which,

if any, scientific assertions are of this sort” (ibid: 27). The aforementioned

extract seems to indicate that Suganami has not heard of ‘approximate truth

claims’ and the fact that the wrongness of a scientific theory comes in degrees.

Suganami concludes that

In the end, I wish to remain agnostic as regards the real (that is, knowledge-

independent) existence of theoretical entities invoked in the currently best

explanation of nature. In fact, I do not understand why it is necessary to go beyond

simply acknowledging that a particular entity is invoked in the currently best

scientific explanation and that implicit in the idea that this entity explains nature is

the supposition that it exists independently of our knowledge. (ibid: 27-8).

A careful examination of the aforementioned claim reveals that it begins with

Suganami being a constructive realist, but ends up with Suganami as a full-

blooded scientific realist. Similarly, he insists that he remains agnostic with

respect to the existence (or not) of unobservable entities, when he takes issue

with the notion of causation in Wendt’s IR theory, in order to argue that all we

need to do “is to point to the causal potentials of collectively held ideas, to

208

which the discussion of the reasonableness or otherwise of the belief in the

knowledge-independent existence of certain subatomic particles would seem to

be irrelevant” (ibid: 30). With regards to this argument I am aligned with Gofas’

criticism “that remaining agnostic on the subject is a luxury we cannot afford”

(Gofas 2007: 70), mainly because “the criterion we invoke for establishing the

reality of unobservable entities is decisive for the question of where does

causality reside” (ibid: 70).

Moreover, I do not think that Suganami has properly understood some

crucial components of CR. For instance, with respect to causation, which

consists in constitutive explanation, Suganami claims that “in short, a causal

relationship has to do with mechanistic coming about or intentional bringing

about of an effect-event (under a relevant description), whereas a constitutive

relationship has to do with making a particular description of something

logically possible” (Suganami 2006: 67). This description of constitutive

explanation is inaccurate, as can be seen by the analysis on generative

mechanisms in Chapter 3. However, because it is an issue which is important

for the study of structures in IR, it is important to highlight the main differences

between a causal relationship and a constitutive one. Yalvac (2010: 170) quotes

Sayer (2000: 11) in noting that “Bhaskar makes a distinction between three

levels of reality: the actual, the empirical and the real”. Moreover, “given the

above distinction, events are caused by structures coming into different

‘contingent relationships’ with each other ‘to co-determine the occurrence of

209

events, thus breaking the one to one relationship between structures and events

found in the closed system of scientific experiments’” (Yalvac 2010: 170).144

As mentioned above, the aforementioned distinction between the ontologies

of empirical realism, constructivism and SR applies to the study of structures in

IR. The empirical realism of both realist and neo-realist IR theories enables us

to consider states as “observable objects (or units) with directly given material

interests that they seek to advance as relatively unified subjects against other

states (units)” (Jessop 2010: 187). The social constructivism of IR constructivist

theories does not only take on board that “social practices are constitutive of

social relations, but also commits the ‘epistemic fallacy’ in assuming that

reality corresponds to the knowledge we have about it” (ibid: 187). Then, SR

makes the difference in that

It…endorses a modified social constructivist position in recognizing that the social

world is always-already meaningful but qualifies this through the claim that there

are complex material as well as discursive mechanisms that shape the variation,

selection and retention of ideas, concepts and practices. This implies that, while all

ideas are equal, some are more equal than others. (Jessop 2010: 187)

If one were a critical realist and would like to make a general comment on the

nature of causation in terms of CR and its ramifications for IR theory, one would

have to argue, along with Kurki, that

144 A similar, perhaps even clearer, description of the ontology of SR, from the perspective of CR, is given by Jessop: “In contrast, SR is committed to a depth ontology that distinguishes in general terms the levels of the real (causal mechanisms that include tendencies, counter-tendencies, capacities, liabilities, etc.), actual (the product of the interaction among a plurality of mechanisms in specific conditions) and empirical (observations concerning the actual and/or the real)” (Jessop 2010: 187).

210

In the case of causation, critical realism, through its particular non-positivist

insistence on the ubiquity of causal forces in natural and social spheres, challenges

positivist assumptions on science and causation in IR and directs IR theorists

towards more holistic, reflective and methodologically pluralistic causal theories. It

also forces post-positivists away from simplistic rejection of the language of

science and causality and, in fact, points out that many of their analyses provide a

way forward for IR theoretical causal analysis. (Kurki 2007: 377-78)145

Returning to Suganami, one might claim that, despite the fact that his comments

on SR are based on a misunderstanding of its premises and an ignorance of the

context within which it has been deployed in the philosophy of science literature,

he does succeed in his critique of the use that Wendt makes of SR in defending

the belief in the reality of the state. Suganami argues that it is unnecessary to

invoke SR in order to defend the argument that states really exist, since in order

to do that all that Wendt needs to do is to point to the “causal potentials of

collectively held ideas” (Suganami 2006: 64)146

Moreover, Suganami is right in arguing that Wendt in his attempt to prove

the existence of a constitutive relationship between social structures and the

agents’ identities and interests - and thus disprove both individualism (the

alleged method of neorealism) and holism (the alleged method of the world-

system theory) as appropriate methods for IR theorizing - comes up with

nothing more than a number of causal narratives. His concluding comment that

Wendt “could not have done otherwise” (ibid: 69) is not to be bypassed, given

145 On the issue of causation in terms of CR and its implications for IR theory, see also Kurki (2006) and Whitham (2012) but, first and foremost, Kurki (2008). Furthermore, for a defence of the need for a multicausal analysis in the study of foreign policy, which is basically rooted in CR’s ontology, see Eun (2012). 146 However, one may argue that this applies just as much to the idea of the state itself, as to that of international society, etc.

211

that Wendt has not yet provided us with a substantial empirical research

proposal, which could lead to “novel verifications” of his own theory of

causation. Furthermore, according to Suganami (ibid: 71), another weak spot in

Wendt’s theory is that it is permeated by an ambivalence as to whether it will

concede to individualism, via the claim that unlike rocks, human beings are

intentional agents that “exist partly in virtue of their own thoughts” (Wendt

1999: 181), or not. Overall, it can be said that on the one hand Suganami has

failed in his critique of SR from a philosophical point of view but, on the other,

he has made sharp-witted comments on SR’s application to substantive IR

theorizing.

Stefano Guzzini and Anna Leander have not grappled with SR very

systematically. Nevertheless, they have made the valid point that Wendt’s

version of positivism - a “sophisticated positivism” as Wendt himself designates

it - shares, notwithstanding Doty’s claim about the opposite (Doty 2000), almost

nothing with positivism (as it is usually understood in IR), apart from the

assumption that there is in principle no difference between the natural and social

sciences. This “sophisticated positivism” holds that all observation is theory-

laden and scientific theories are tested only against other scientific theories and

not against the world itself; a situation which renders the search for secure

foundations of knowledge impossible (Wendt 2000: 73). It is important here to

single out the claim by Kratochwil (2006) who has aptly observed that Wendt’s

idea of SR seems not to be compatible with the aforementioned position

according to which one cannot test scientific theories against the existing world

(Guzzini & Leander 2006c: 79, fn. 4). However, Guzzini and Leander argue that

“Wendt’s project is best understood as a grand synthesis than a via media” (ibid:

212

91), since it makes divergent positions (e.g. positivism and interpretivism)

complementary, “by reconfiguring them under a new encompassing framework”

(ibid: 91).147 Therefore, it is even more interesting that they underscore that “it

comes as no surprise that Wendt’s present project is to use the idea of

complementarity in quantum theory – a post-positivist natural science, as it were

– as a basis of social science” (ibid: 91). It is obvious that Guzzini and Leander

believe that the idea of complementarity is a permanent feature of all versions of

the Wendtian IR Theory. As already mentioned, I will deal with Wendt’s

ambitious new project towards ‘a quantum social science’ in chapter 5 but, for

the time being, I cannot help but highlight the fact that quantum mechanics is by

no means a “post-positivist natural science”. Even more so, to compare Wendt’s

encompassing IR Theory with the way Einstein embedded Newtonian physics,

as Guzzini and Leander do (2006c: 73), can be at best described as a bad type of

science fiction.

Furthermore, Jackson (2008a) has some important points with regard to the

distinction that Wight makes between philosophical and scientific ontology.148

In the presentation and discussion of Jackson’s stance vis-à-vis Wight’s CR,

which follows in the next few pages, I come back to some philosophy of science

issues which I have already discussed in chapter 2. This is necessary in order to

show that, although Jackson (2008b, 2011) and Wight (2006) deal extensively

with philosophy of science in their works, they don’t grasp the complexity of the

147 Contra Guzzini and Leander, Milan Brglez claims that, due to his SR, Wendt “rescues” positivism. If he is right, the problem lies in that SR is then incompatible with “thick constructivism” and other post-positivism alternatives. At the end, Brglez argues that, “because of (his) scientific realism, IR theorists are better off engaging (not ignoring) him (Wendt)” (Brglez 2006: 354). 148 In case one is interested in a very detailed analysis and critique of SR by Jackson, they can read chapter 4 (“Critical realism”) of Jackson (2011); cf. Jackson (2008b).

213

relevant discussion as it has developed in the relevant bibliography. This is

something that has crucial consequences for the introduction of SR/CR in IR.

First, Jackson argues that one can ascribe to only one of these two ontologies,

instead of the whole package. To put it differently, if one does not want to accept

both philosophical and scientific ontologies, one can either accept Wight’s

ontology of agents and structures without accepting his philosophical ontology

of a mind-independent reality, or the other way around. Second, as has already

been discussed, Wight’s position is against the ‘epistemic fallacy’ of reducing

what is to what is known (based on the assumption that all we take to exist must

be knowable to the human mind). In Wight’s view, “statements about being

cannot be reduced to, and/or analyzed solely in terms of, statements about our

knowledge of being” (Wight 2006: 28). Jackson (2008a) points out that by

arguing the above, Wight is led to conclude that the existence of unobservable

entities (like quarks or social classes) is “not dependent upon their specification

in any theory” (Wight 2006: 32). However, one should bear in mind that, as it

stems from my discussion in chapter 2, according to the epistemic thesis of

either SR or SSR (Worrall’s moderate version of SR), only entities (both

observables and unobservables) which are posited by mature and predictably

successful scientific theories, or, at any rate, entities very similar to those posited

are taken to inhabit the world. In other words, the existence of entities is

dependent upon their specification in a theory (in contradistinction to what the

Bhaskarian/Wightian SR holds).

This verifies the claim of the previous subsection according to which Wight

prioritizes ontology over epistemology, since he does not commit to any

particular set of posited entities but, instead, it is the empirical research and

214

subsequent evidence that in each distinct case decides about the existence or not

of the posited entities. Jackson also makes the point that if the posited entities

are not taken as objectively existent, as philosophical realism demands, then they

could be simply regarded as “useful analytical postulates”, since in this case one

could leave the question of their existence aside (Jackson 2008a: 342-3).

Wight provides three reasons as to why key theoretical entities and generative

mechanisms should be taken to be real and not merely useful:

(1) SR corresponds to scientific practice better, since scientists organize their

experiments by putting them to the test; in Wight’s own words, “Scientific

realism makes intelligible what scientists do” (Wight 2006: 24). Wight has

made the same point in his “manifesto for SR in IR” where he argues that “the

starting point for understanding SR is to realize that it is, first and foremost, an

account of what scientists actually do in their practice” (Wight 2007: 382).

Chernoff rightly counter-argues that “this is quite far from a philosophical

doctrine of SR” (Chernoff 2007: 401). Jackson’s comments that he is not sure

“why the observation that scientists probe and test their theoretical postulates

necessarily entails realism” (Jackson 2008a: 343) are in the same vein, given

that even anti-realist approaches such as causal conventionalism (Chernoff

2005) can make intelligible what scientists do. Moreover, he holds that “even if

it were incontrovertibly the case that scientists treated their theoretical posits as

if they were real, this would say nothing whatsoever about whether those

theoretical posits were in fact real – or about whether scientific progress in some

sense depended on the search for real-but-unobservable entities and

mechanisms” (Jackson 2008a: 343). A proper examination of the above

215

statements necessitates extensive argumentation in terms of analytic philosophy

and goes well beyond the scope of this thesis. Anyway, there are four initial

points to be made on the above.

First, Wight’s thesis is a rather controversial one. Were it for a realist like

Psillos (1999), he would have claimed that if realism were about science

practice, then it would have been an empirical view and could be therefore

falsified. However, other philosophers of science, like Giere,149 argue that

realism is in fact about science practice and scientists are realists in practice.

One may counter-argue that even this view is problematic, because the practice

underdetermines the philosophical stances. Therefore, it may probably be best to

argue about realism directly. This can be done by arguing, for instance, that

taking theories to be true is the best explanation of their success. What an IR

critical realist like Wight might need here is an account of success concerning

IR theories as well as an analysis of existing theories that are successful.

Second, I would like to draw attention to my presentation of both SR and its

alternatives in the second chapter of this thesis. The arguments presented there

allow me to claim that apart for Chernoff’s conventionalism – which draws on

Duhem ([1908] 1969) - there are also other anti-realist alternatives to SR.

Consequently, it is uncertain why one would have to choose Chernoff’s

conventionalism instead of one of its alternatives.150 I personally believe that

those social scientists, including IR scholars, that are interested in the

philosophy of science, but do not adhere to the positivist epistemology, are

149 I have referred to his works (Giere 1998, 2005) in chapter 2. 150 In Ellie Zahar’s Poincaré’s Philosophy: From Conventionalism to Phenomenology (2001), “Appendix IV: Ramseyfication and Structural Realism”, which is written by Worrall and Zahar (Zahar 2001: 236-251), contains the conventionalist ideas on which Worrall has mainly based the argumentation which he has presented in his most recent and updated work on SSR (Worrall 2008).

216

rarely enchanted by the role Mathematics play in the structure of scientific

theories. Moreover, they do not investigate its philosophical significance and, as

a consequence, have never paid any attention to Poincare’s Science and

Hypothesis ([1905] 1952), as, on the contrary, Zahar and Worrall (2001) have

done. Instead, they usually (and almost unavoidably) draw on the general

philosophy of science literature, and do not try to follow technical philosophy of

science works on QM, General Relativity or Quantum Field Theories. The

latter, however, presupposes an understanding of the mathematical frameworks

of these theories, and since this is the case, it is not accidental that it is in this

kind of works – see, for instance, Cao’s From Current Algebra to Quantum

Chromodynamics (2010) and Ruetsche’s Interpreting Quantum Theories (2011)

– where one can trace a rather strong influence of Worrall’s SSR. At this point,

one may plausibly wonder about the value of familiarity with the advanced

knowledge of Mathematics and Physics which is required in order to understand

technical philosophical works which are concerned with the foundations of

physical theories that are at the very forefront of both theoretical and empirical

research in natural sciences. The answer is that normally one should not be

preoccupied with them. However, such a preoccupation becomes necessary,

from the moment that some philosophically inclined IR scholars, like Wight,

suggest that IR should endorse SR as its meta-theory, since it describes more

accurately the way scientists work (and the way social scientists ought to work).

One must point out that other IR scholars, like Jackson, disagree and argue that

Chernoff’s conventionalism could probably explain better than SR how in fact

scientists work. With respect to the above, I am uncertain on whether

philosophy of science can explain better than science itself how scientists work.

217

However, if one finally (be it right or wrong) resorts to the philosophy of

science in order to address this question, then one must be familiar with

technical philosophical works such as the aforementioned ones. If this is not the

case, it is difficult to have a comprehensive picture of what SR or SSR or

different anti-realist alternatives stand for.

Third, I think Chernoff makes a good point when he argues that “Wight’s

preference that scientists hold ‘realist beliefs’ about the unseen world of

theoretical entities is an example of what Arthur Fine calls ‘motivational

realism’; it neither implies nor is implied by SR” (Chernoff 2007: 402).

Fourth, I must accept that CR seems compatible with the type of entity-

realism, which ties more with experimental practice and avoids the question of

theoretical truth. However, in chapter 2, I argued that this position would be

difficult for a realist to adopt, since it is theories that provide the realist with

knowledge of any properties these entities have and allow for the interpretation

of experimental results.

(2) Taking theoretical entities to be real motivates further empirical

investigation, in the sense that “if scholars are only using theoretical terms, like

‘structure’, instrumentally, then ‘there is little need for them to make clear how

to use the term’ since the only issue is whether the ‘postulated term helps

explain/predict the phenomena’ and not whether the term accurately captures

anything really existing (Wight 2006: 122)” (Jackson 2008a: 343). My first

comment is that this issue belongs to the huge discussion about the status of

theoretical entities in both SR and its anti-realist alternatives and is almost

identical with the one I have made just a few lines above. My second comment

218

is that Jackson rightly uses the following good example from IR theory, in order

to argue that even philosophical non-realists assign specific meanings to their

terms. According to Jackson:

…it would be a stretch to argue that, for instance, a philosophical non-realist like

Waltz doesn’t make clear what he means by ‘structure’; the fact that Wight spends

many pages arguing with Waltz’s conception of structure would seem to support

the point. So while it is clear that a philosophical realist has to conduct

investigations into their posited theoretical entities, it is not clear that a non-realist

would not do so – albeit on different grounds, such as the demand for logical

clarity and precision. (Jackson 2008a: 343)

However, Wight is so convinced of the solidity of this argument because if it

does not hold, a number of questions remain unanswered. Specifically:

How can we put theory into practice if our theoretical posits do not exist? What

does not exist cannot do anything; possess causal power, bring about change, or be

changed. And why should we bother to expose theoretical assumptions underlying

practice if we deny the existence of those same assumptions? (Wight 2007: 381)

There are a number of points which can be made on the aforementioned. If

prediction is what really counts, then it is not clear why theoretical assumptions

should be true. So the realist needs to argue that a) explanation counts and b)

explanation requires truth. Anti-realists, on the other hand, argue differently.

According to them, we need theories not because we want true pictures, but

because we need systematization and classification. Even theoretical

219

assumptions can be criticized from this perspective for being more or less

useful, plausible or convenient. The conclusion is that more argument is needed

here in defence of realism.

Here is one more example of the extent to which Wight emphasizes this

point:

At the heart of the SR account of science is the view that the entities

postulated by mature scientific theories (electrons, genes, viruses, dark

matter, black holes etc.) are believed by scientists to be real. If scientists did

not believe them to be real how do we explain attempts to split atoms,

provide cures for viruses, or the search for dark matter? (Wight 2007: 382)

If one wishes to take a more critical stance on Wight than the one I have taken,

one could argue that the above thesis is a descriptive one, and as such, it might

be wrong. Realism is a view about what exists as well as how that exists (mind-

independently, etc.); the latter question being definitely applicable to natural

sciences. Now, the problem of what mind-independence in the social sciences

consists of is a serious one,151 but I think there are ways to construe it. For

instance, Psillos, in his paper “Scientific Realism and Metaphysics” (2005),

provides us with a framework within which such a construal becomes

legitimate. He argues that there are two ways of conceiving of reality; either as

comprising all facts or as comprising all and only fundamentalist (meaning

irreducible, basic) facts. Psillos argues that SR should be committed to the

factualist view of reality (not to the fundamentalist one), an assumption which

151 For example, if diplomats do not believe in the existence of an international system, as Mrs Thatcher did not believe in ‘society’, it has major consequences.

220

he relies upon in order to develop a series of arguments before his final

conclusion that SR “is independent of physicalism and non-Humeanism and that

the concept of truth is required for a sensible understanding of the metaphysical

commitments of SR” (Psillos 2005: 385). I believe that, for a realist, it is within

this very theoretical framework that the examination of the issue of mind-

independence in social sciences might become more reasonable and feasible.

Furthermore, following Wight again, one notices that

It is not only the practice of science which validates a realist treatment of

theoretical terms. For we use this knowledge to manipulate these entities in a

practical sense. (ibid: 383)

This is a crucial argument that needs further elaboration. As the argument

stands, it might make sense in physics and perhaps psychology, but if it is to

make sense in IR it is going to need development, defence as well as some

examples from IR itself. What are the relevant ‘social experiments’? And again,

how broad are the conclusions from these ‘social experiments’? Do we conclude

the truth of the relevant part or just the existence of the relevant entities? These

are three decisive questions which must be addressed in order for the above

argument to be evaluated within the field of the social science of IR. Otherwise,

without these clarifications, Wight seems to waver between the existence of

entities and the truth of theories.

It is worth noticing here that all the above comments I have made on Wight’s

statements with respect to the status of the entities of scientific theories, are

221

comments to some of which a ‘traditional’, so to say, scientific realist (not a

critical realist) could have presumably ascribed to.

To sum up, Wight’s emphasis on his argument that one should take

theoretical entities to be real is not well justified, since it needs further

elaboration. The latter is not an easy task, for things are rather unclear with

Wight’s CR. On the one hand, it endorses the scientific realist assumption that

theoretical entities should be considered to be real, which is compatible with the

other scientific realist assumption that there exists an independent world out

there. On the other hand, CR’s commitment to the idea of rejecting the

‘epistemic fallacy’ of reducing what is to what is known, along with his heavy

prioritizing of ontology over epistemology, ‘force’ one to conclude that Wight’s

epistemic-relativism leads to a truth-relativism. Indeed, as I have clarified in the

former subsection, where I presented the basic features of Wight’s ASIR, CR

endorses a truth-relativism. Finally, the following passage from Wight’s

“manifesto for SR” is most illuminating:

Given the historically specific nature of knowledge, we have to accept the fact of

epistemological relativism; the view that all beliefs are socially produced and that

neither truth values, nor criteria of rationality, exist outside of history. ‘Whenever

we speak of things or of events, etc. in science we must always speak of them and

know them under particular descriptions, descriptions which will always be, to a

greater or lesser extent, theoretically determined.’152 But epistemological relativism

in this sense does not imply judgemental relativism, and we can, and do, make

rational choices between competing knowledge claims. (Wight 2007: 386)

152 Bhaskar [1975] 2008: 249.

222

(3) Only if theoretical entities (e.g. such structures as ‘social classes’ or ‘global

capital market’) are taken to be real, can social science unveil their impact on

the formation of oppressive social relations and thus fulfill its emancipatory

role. Jackson (2008a) wonders if this reason is of any importance to those IR

scholars that, like himself, do not assign a socio-political role to any IR theory

or meta-theory, because they “do not believe that the goal of social science is to

contribute to human emancipation, or even to carve out a special place for

human beings by identifying ‘properties best reserved for human agents’ (Wight

2006: 206)?” (Jackson 2008a: 344, italics are mine).153 Although I consider

myself to belong in the same (Weberian) tradition as Jackson and, consequently,

would not like to assign any politically progressive/emancipatory role to social

sciences, I must admit that critical realists are not completely wrong in their

argument that meta-theoretical choices are not inconsequential but, on the

contrary, have political connotations and implications for the substantive IR

theories proper (see, for instance, Wight 2006, Kurki 2009, 2011 and Kurki &

Sinclair 2010). These implications are overestimated, but they should not be

totally ignored. One moderate and balanced argument about the alleged

progressive political implications of critical realists’ emphasis on ontology is

due to Kurki. Kurki argues that statements of the kind that critical realists “point

out that positivist assumptions about like-law regularities reify the social world,

making present social systems natural” (Kurki 2009: 450) maybe rather

problematic a reading, for “there is no singular political interpretation of

positivism”. However, it is telling that she adds that “yet, this does not mean

153 For a reply to Jackson (2008a), see Wight (2008), although, strangely enough, this time Wight is as not analytical in his counter-arguments as usual.

223

there is no politics arising from positivism, or that the political charge of

positivism has not been important for positivists” (ibid: 450).

Richard Ned Lebow (2011) provides a critique of SR, which is developed

from a different perspective to that of Jackson’s. Lebow’s difference comes,

first, from his anxiety about whether SR can tell us anything about the nature of

human behaviour, i.e. the human condition,154 and, second, from his anxiety

about whether social entities differentiate themselves from the natural ones in

terms of substance.

Lebow is correct in his argument that the end of Cold War has brought about

the dethroning, on the one hand, of neorealism from its position as the

overwhelming IR theory and the emergence, on the other, of classical realism

and constructivism. Classical realism has undergone a renaissance after the

collapse of communism because of the hegemonic behaviour of the world’s

leading nation-state and the ethnic, cultural and religious conflicts that have

recently surfaced in many places throughout the world. Constructivism is also

reinforced in the IR system of ideas because of the consensus within the IR

community that the ideas of liberty and human rights have played a decisive role

in undermining the fabric of the former Soviet Union and that questions of

identity appear to be central to many post-Cold War conflicts. This view is

supported by liberalism and the English School, which claims that ideas are

important for the shaping of international politics. Despite the above, it is

constructivism which is mainly associated with the aforementioned theoretical

arguments, for constructivism probably possesses more appropriate tools in

explaining the formation of collective identities than any other relevant rival

154 See Hanah Arendt’s, The Human Condition (1999), where the author reflects on problems of human agency and political action.

224

theory. This is the reason constructivism has been given more credit for its

ability to accurately explain current international affairs than that given to

liberalism and the English School.

Lebow reminds us that classical realism “is rooted in the tragic understanding

of politics associated with the writings of Thucydides, Machiavelli and Hans J.

Morgenthau” (Lebow 2011: 1220) and makes the following acute statements

with regards to the basic features of classical realism:

Like Greek tragedians, classical Realists tend to regard history as cyclical; efforts

to build order and escape from fear-driven worlds, while they may succeed for a

considerable period of time, ultimately succumb to the destabilizing effects of

actors who believe they are too powerful to be constrained by law and custom.

Classical Realism stresses sensitivity to ethical dilemmas and the practical need to

base influence, wherever possible, on shared interests and persuasion. (ibid: 1220)

Lebow (2008) offers a constructivist theory of political order and IR, which is

based on theories of motives and identity formation that derived from the

ancient Greeks and insists that the role of agency cannot be properly understood

through the singular study of the agent-structure relationship (which is

underpinned by SR in Wendt and Wight). On the contrary, he claims that

“philosophers and social scientists have attributed an array of often

contradictory traits to human beings, and at least as far back as Thucydides,

more sophisticated thinkers have suggested that we are a grab bag of conflicting

225

tendencies” and that “circumstances appear responsible – at least to a degree –

for determining which tendencies come to the fore” (ibid: 1226).155

In his most important objection to SR, Lebow deploys his ideas about the

nature of social ‘facts’ along the lines of the Weberian tradition to which he

adheres. For Lebow:

Social ‘facts’ are reflections of the concepts we use to describe social reality, not of

reality itself. They are ideational and subjective, and depend on other equally

subjective concepts, never making contact with anything real in the sense that

temperature does. Even the existence of reasonably precise measures for social

concepts – something we only rarely have – would not make them any less

arbitrary. (Lebow 2011: 1227-28)156

Next, Fred Chernoff, who has a solid background in the philosophy of both

natural and social sciences, as is evident in the coherence and clarity of his

argumentation in his paper, “Scientific Realism as a Meta-Theory in

International Politics” (2002), argues that

Wendt offers two sets of considerations motivating adoption of SR. His foremost is

roughly: 1. Meta-theory should not rule out substantive theories we might 155 It should be, however, considered whether recent works in cognitive psychology, such as Pinker (2011), could be taken on board by IR theorists who attempt to explain and understand agents’ actions in international politics. In the aforementioned book, Pinker argues that, contrary to popular belief, over millennia and decades, humankind has become progressively less violent. He attributes this to the power of progressive ideas; modernity and its cultural institutions make us better people and reduce violence within and between societies. Pinker’s argument in this book is contested - see, for instance, the book-review by Micale (2012) - but I think that the ideas it brings into the discussion about violence, including warfare, are of interest to IR scholars and may constitute a good future research program in IR. In any case, given our greater capacity for mass murder and evident willingness to implement it on occasions, one may doubt whether it is meaningful to argue that humankind has become progressively less violent. 156 One may note here that these social concepts become real in the sense that we act as if they were facts.

226

otherwise accept. 2. SR has no such effect, but anti-SR does. 3. We should,

therefore, accept SR as our meta-theory. This paper has tried to show that the

argument is unsound because both premises are false. The first premise is false

because we should have some a priori criteria for a properly formed theory, which

should at least include requirements like “internal consistency”. The second is false

because SR does require the assumptions be “realistic”, and avowed

instrumentalists need not accept this requirement. (Chernoff 2002: 205).

However, in order to do justice to Wendt’s argument that SR is compatible with

methodological pluralism – which implies that SR does not rule out differing

substantive IR theories – I must say that the first of Chernoff’s points is rather

unjust; namely, there is no reason to conclude that Wendt is willing to accept

any IR Theory that does not fulfill the criteria of logical coherence, internal

consistency, etc. Chernoff’s second point is valid although, as I have discussed

in the second chapter of the thesis, the debate between scientific realists and

anti-realists requires a rather complex and sophisticated argumentation as to

what it means for their assumptions to be ‘realistic’.

Chernoff continues as follows:

Wendt’s ancillary argument motivating SR is that IR is capable of more progress

that has been exhibited because there is unnecessary disagreement over

epistemology. He seems to reason as follows: 1. Agreement on foundational

questions aids progress in the discipline. 2. Most IR scholars accept SR, even if

they do not acknowledge it openly. 3. It follows from 2 that there is less

disagreement among IR scholars about ontology than about epistemology. 4.

Therefore, a shift from epistemological to ontological foundations aids progress.

This second argument is invalid because premise 3 does not follow from premise 2.

227

Even if IR scholars accept SR, it does not follow that they share ontological views,

since SR merely states connections between a theory and its ontological

commitments; it does not commit IR theorists to any particular ontology (because it

does not motivate adoption and any specific substantive theory). (ibid: 205)

Here Chernoff makes a simple, albeit crucial argument, which I would like to

emphasize once more since I have the impression that it is very often

overshadowed by the discussion over the specifics and technicalities of both SR

and its alternatives: SR does not commit IR theorists to any particular ontology,

thus enabling the adoption of differing substantive theories. What does this

mean in practical terms? Which entities’ existence is going to be tested

empirically, at the end of the day? From the point of view of the kind of

conventionalism that he defends, Chernoff (2007) questions Joseph’s (2007)

emphasis on the value of theoretical entities’ emergent properties in terms of

their ability to help us explain what we observe around us:

The particular emergent properties postulated by, say, one version of structural

Marxist theory differ from those postulated by other Marxist variations and by non-

Marxist theories. Which, then, are the ‘real’ emergent properties? If all competing

theories ‘talk about’ certain structures and properties, they are not thereby real. We

often have minimal grounds for selecting one theory over the rivals and is a

questionable move to insist, as scientific realists do, that the entities referred to by

one theory exist, while all others do not. (Chernoff 2007: 405)

Chernoff makes a number of interesting comments on various aspects of SR.

One of them is about the validity of the application of the No Miracle

228

Argument157 in the social sciences, which he rejects since he argues that the No

Miracle Argument is fit for the explanation of scientific progress of only the

natural sciences (Chernoff 2007: 404).

Finally, Chernoff (2002) qualifies Quasi-Duhemianism (henceforth QD)158

for a possible meta-theory of IR but I am not going to present QD here, since the

aim of this chapter is not to provide any alternative to SR meta-theory, but only

to analytically discuss the pros and cons of the introduction of SR into IR. For

the time being, suffice it to say that I am aligned with Jackson who goes for a

‘pluralist science of IR’. The latter “posses the challenge of dealing with bodies

of warranted knowledge stemming from philosophically incompatible

methodologies” (Jackson 2011: 210, italics are mine) and relies on the key-

concept of the ‘translation’ of each one of these methodologies into the

languages of the others:

Instead, the implication of methodological pluralism is that between different

bodies of warranted knowledge we have the ongoing challenge of translation:

literally, the task of making claims comprehensible to speakers of other

methodological languages…methodological pluralism sets up a variety of

contentious conversations and efforts to appreciate the insights of alternative ways

of producing knowledge while avoiding the temptation to universalize our own

methods of conducting scientific inquiry.” (ibid: 210)

157 I have discussed the No Miracle Argument (NMA) in Chapter 2. 158 In a number of successive papers Chernoff comes back to the issues of SR and QD (Chernoff 2007, 2009, 2009a). However, the basic line of his argumentation about SR and QD is the one he has presented in his 2002 paper. To cut a long story short, I agree with Chernoff that “the merits of Wendt’s work are to be found in the substantive theory” (Chernoff 2002: 205), not in its meta-theory.

229

It is questionable, however, whether one needs such an extensive argumentation

on meta-theoretical issues in order to claim that methodological pluralism is the

right approach to the study of IR and to acknowledge that each methodology

must understand what the alternative methodologies really do and predictably

result to. Other IR scholars have come to the same conclusion in a much more

direct way, namely by just observing the developments in real-world politics

and invoking the appropriate theories in order to capture different aspects of

complex political phenomena. For instance, Christopher Hill and Michael Smith

adopt methodological pluralism as one of the three methodological assumptions

they make in order to study EU’s international relations, when they argue:

It follows that a methodological pluralism is therefore required when seeking to

explain and understand the EU’s international relations. No one approach,

whether broad-brush as in realist, rationalist and constructivist, or more specific, as

in geo-politics, intergovernmentalism or ‘expectations’, comes near being adequate

by itself. The usual problems then arise of how to relate diverse, and possibly

incommensurable middle-range, theories to each other, but these are inevitable in

any attempt to do justice to complexity. (Hill & Smith 2011b: 8)

Hill and Smith not only defend methodological pluralism but also put on the

table the inevitable question about how one could “relate diverse, and possibly

incommensurable middle-range, theories to each other”; which is exactly the

problem of the ‘translation’ of one methodology to another that Jackson has also

put forward.

As for Jackson’s ‘pluralist science of IR’, Onuf wants it to be even more

pluralist, in the sense that Jackson concedes ground at positivism at the start of

230

The Conduct of Inquiry in International Relations (2011) by aspiring to science

as a model of scholarship. Onuf argues that the obvious reason this should not

have happened is positivist resistance to the language turn. Indeed, no social

theoretical framework today “can dispense with the nuts and bolts of language

to hold it together and still be inclusive” (Onuf 2012: 26); for him, it is

noticeable that “language and speech have no entries” in Jackson’s index (ibid:

26). Had Onuf claimed the opposite, he would have renounced his own major

contribution to the field, namely his constructivist IR theory (Onuf 1989). There

are, of course, many other well-known IR theorists who have spelt out their

disagreement, albeit on different grounds, with the adoption of science as a

model of scholarship for IR. Two such theorists are Hans Morgenthau (1946)159

and Hedley Bull (1966).160

With regard to the development of CR as a science of IR, Chris Brown

(2007) argues that one should not ignore the political roots of Bhaskar’s work.

In fact, Bhaskar ([1975] 2008) reacted against the disillusionment with science

in the 1960s and 1970s – a symptom of which was the decreasing interest in the

‘scientific Marx’ of Capital that the radical intellectuals of that epoch have

shown. The latter is, according to Brown, “the ur-text of critical realism”

(Brown 2007: 410). With this work Bhaskar aimed to put “new life into the

materialist approach to social theory, and to combat the idealist leanings of

1960s radicalism” (Brown 2007: 414). Brown finds that “the kind of Marxism

that Bhaskar and his followers wished to re-legitimate in the face of criticisms

of 1960s romantics is hard-headed and a little cold-blooded” (ibid: 416).

Although not a Marxist himself, Brown calls for more works like that of Joseph 159 For a reconsideration of Morgenthau’s work, see Williams (2007). 160 For a defence of the scientific approach to IR, see Kaplan (1966) and Nicholson (1985), (1996) and (1999).

231

(2007), with the thought that they may lead to a genuine revival of historical

materialism in IR studies. He makes this point when arguing that

Reading Joseph, it is clear what we are being offered – critical realism is not

simply a negative movement saying what is wrong with existing theories, but

provides the positive basis for another and – allegedly – better theory: namely

Marxism (Brown 2007: 415).161

Τhere are, indeed, some Marxist IR scholars who use CR as the meta-theoretical

foundation for their work, in the sense that they combine aspects of SR with

elements of Marxist theory, in order to refine the latter and to reawaken the

interest of the IR community in the utility of Marxism for the study of the

international.162 Indeed, Marxism could be probably fertilized through CR, due

to the latter’s emphasis on ontological depth. This could lead to the integration

of the economic theory of Capital and the Marxist theory of historical

materialism. Such a development might provide us with a Marxist theory of the

international, namely an IR theory in the categories of capitalist economic

evolution, which is now lacking (cf. Υalvaç 2010).

For the above reason, though not a Marxist myself, I agree with Brown’s

suggestion that we need a resurgence of interest in Marxist studies within IR

theory. Furthermore, I strongly agree with him that if the effect of CR is to

161 Wight and Joseph replied to Brown (2007) and distinguished SR/CR from Marxism (Wight & Joseph 2010: 3-4; cf. Lebow 2011: 1221). They are right to argue that SR/CR is not reducible to Marxism because, among other reasons, SR “draws on a wide, and differing, range of intellectual resources that go well beyond Bhaskar” (Wight & Joseph 2010: 3). That notwithstanding, I am aligned with Υalvaç in that “CR is compatible with diverse theoretical positions as a philosophical project, but it is most compatible with Marxism” (Υalvaç 2010: 185, emphasis added). 162 See, for instance, Jessop (2001), (2002), (2007), (2010), Joseph (2002), (2007), (2008), (2010a), Υalvaç (2010), etc.

232

“revitalize debates over epistemology and ontology it will do the discourse no

service” (Brown 2007: 416).

In terms of the discussion on the role of science within IR theory, now, I

would also like to call attention to John Lewis Gaddis’ paper, “History, Science

and International Relations” ([1996] 2002). Starting from the idea that historians

and social scientists suffer from ‘physics envy’ or any other kind of

methodological ‘envy’, he comes to demonstrate that even ‘hard sciences’ such

as geology and paleontology are ‘historical sciences’, “whose strength lies in

explaining in great detail where we are and how we got there, but whose

pretensions to forecasting are confined to much less specific observations about

the overall framework within which certain known processes will occur”

(Gaddis [1996] 2002: 39, emphasis added). Thus, among others, he emphasizes

the importance of rediscovering the importance of the historical narrative,

which may well serve as “a kind of bridge between the ‘new’ hard sciences of

chaos and complexity and the ‘old’ social sciences” (ibid: 45). He does this in

order to argue that “by sticking with narratives, the historians who never bought

into the ‘old’ social sciences in the first place have achieved something rather

remarkable: they have come out on the cutting edge of a revolution by persisting

in a reactionary stance” (ibid: 45, italics are mine). When he reaches the end of

the road he argues that a rediscovery of narrative by social scientists “could

move us back towards the pre-professional era when intelligent people could

comfortably involve themselves in, and learn from, multiple disciplines without

being regarded as dilettantes” (ibid: 45).

I believe that we need both approaches for the study of IR; that is the

scientific one (whether it confines itself to positivism or accepts what is meant

233

by Jackson’s spacious ‘pluralist science of IR’) as well as the non-scientific

ones (which include the historical narrative). As a consequence, I think that IR

scholars and social scientists in general, may ‘envy’ Physics the way

philosophers ‘envy’ Mathematics. Indeed, as Gian-Carlo Rota, a distinguished

mathematician and philosopher has argued in his paper, “Mathematics and

Philosophy: The Story of a Misunderstanding” (Rota 1990), the attempts of

certain philosophers in the twentieth century to mimic the language, the method

and the results of Mathematics have harmed Philosophy. Even more so,

according to Gian-Carlo Rota, one could go as far as to say that the

aforementioned attempt resulted from a misunderstanding of both Mathematics

and Philosophy and has, in fact, harmed both disciplines.

To conclude, I would argue that the fact that SR has been uncritically

introduced in the study of IR as its proper meta-theory - the examination of the

arguments against SR as a philosophy of science has been rudimentary, if not

entirely lacking - has led to an extensive but perplexed argumentation about SR

within IR. Things are even more unnerving when some IR scholars cannot

distinguish between SR and CR. The result is that the debate over SR as the

proper IR meta-theory is not productive and, even worse, cannot help the

building of substantive IR theories in a substantial manner. This does not,

however, devalue completely the general consideration of and reflection on

meta-theoretical issues.

4. 4 Two important issues raised by the SR debate in IR

During the SR debate in IR a series of interesting issues have surfaced, the

majority of which I have discussed in the second section of this chapter. Here, I

234

will discuss two of them which are of a quite different nature from one another:

the first one is the causes vs. reasons issue, which belongs to the subject matter

of the philosophy of language while the second one has to do with the so-called

‘meta-theoretical hypochondria’, from which IR theorizing allegedly suffers.

4.4.1 Causes vs. Reasons

The introduction of the constitutive explanation (which asks questions of the

kinds “how-possible?” and “what?”) as a complement (not an alternative) to

causal explanation (which ask questions of the kinds “why?” and to an extent

“how?), inevitably poses the question about whether reasons can be causes. In

his paper, “On constitution and causation in international relations” (1998),

Wendt argues that it is wrong to take that material conditions (pace natural

world) imply causal theorizing whereas ideas (pace social world) imply

constitutive theorizing. This is because he suggests that ideas, in the form of

reasons, can be causes (as in rational choice theory) and clarifies that “in saying

that reasons can be causes I am taking one side in a debate about what remains a

controversial issue; for an opposing Wittgensteinian view see…” (ibid: 107, fn.

18). Steve Smith’s comment on this point is that he is concerned that “an

idealist account seems best suited to exactly this alternatively Wittgensteinian

account, rather than the one that stresses the ultimately causal nature of the

social world” (Smith 2000: 153, emphasis added). Smith defends the distinction

between explaining and understanding (Hollis & Smith 1990) on the basis that

explaining is the appropriate epistemology to the natural world, whereas

understanding is the appropriate epistemology to the social world. As a

consequence he also defends the distinction between causes and reasons.

235

Furthermore, Smith makes the valid point that, contrary to Wendt, two other

leading constructivists, Kratochwil and Onuf, consider reasons to be only

constitutive. Kratochwil argues “that there is a crucial difference between causal

explanations in the world of observational facts and that of intentions”

(Kratochwil 1989: 25), while “Onuf’s version of constructivism, like

Kratochwil’s, stresses the rule-based nature of social life, and paints a far more

nuanced conception of the nature of regulative and constitutive rules than does

Wendt” (Smith 2000: 159-60).

Now, if one wants to deal in some considerable depth with the reasons vs.

causes issue, one should first address the question: Can the reason why one acts

be a cause of his acting? It would seem that this depends on the answers to two

different questions: (a) Are actions caused? (b) Might the reason one gives for

the way one acts, actually be the cause of one’s actions? It is possible that the

first question belongs to behavioural science; the second one, however, calls for

an analysis of the logical status of the notion of explanation. One wants to know

whether giving reasons is equivalent to producing causal explanations.163

Two approaches have crystallized from the vast philosophical literature on

this subject. At one extreme, one might hold, like Blackburn (see 1994: 321),

that intentional behaviour is explained teleologically by reference to an agent’s

reasons (beliefs, intentions, wants):

At the other end, one might simply identify, like White does ([1968] 1977:

17), reasons with causes, subsuming both of them under the concept of events

163 It is clear that if we want to know why someone acts as they do, surely we need to know both their intentions and the pressures on them from outside circumstances. The latter is what John Vincent, following Burke, called ‘the empire of circumstance’ (Vincent 1994; cf. Hill 1996b).

236

related by causal laws. In this case, reasons are assimilated to causes because of

the similarities between the two kinds of explanation of action:

The first viewpoint — reasons must be distinguished from causes — stands

in the tradition of the hermeneutical distinction between the explanation

(Erklären) of the natural sciences and the understanding (Verstehen) of the

social sciences. We find its best expression in the philosophy of the later

Wittgenstein, at least as developed in the works of his more ‘orthodox’

followers, G.E.M. Anscombe, Peter Winch, and, more recently, P.M.S. Hacker

(see Glock 1996: 75).

In the social sciences, Peter Winch's The Idea of a Social Science and its

Relation to Philosophy stands out for insisting that the notion of cause is very

different in the natural from what it is in the social sciences. According to this –

standard – interpretation, Wittgenstein himself eschewed a uniform,

nomological paradigm of causation,

which so fascinates us when philosophizing. We are tempted to subsume all cases

under this one paradigm, despite the fact that, as we all know (even if we often

need to be reminded of the fact), there are countless correct uses of ‘cause’ and

related causative verbs that do not fit this special paradigm. Specifying a cause is

one category, itself logically diverse, of explanation of change. Specifying a

reason, a person’s reason, for believing something or for doing something, is

another. (Hacker 1996: 158)

This distinction, however, has been severely criticized by Davidson (1963),

who occupies the contrary position. Davidson claimed that the existence of a

reason is a mental event, causally linked to the acting; otherwise, acting and the

237

reason for which the action is performed would be loose and separate. The

reason that explains an action, then, is the one that was causally efficacious in

prompting the action. Davidson concludes that although we explain action by

reference to reasons, these are causes that are identical with (though not

reduced to, but possibly supervenient on) neurophysiological phenomena.

Apparently, the picture that motivates Davidson is a causal conception of the

mind according to which, mental phenomena are the inner causes of outward

behaviour (see Glock 1996: 75).

Wittgenstein’s distinction stands at odds with this picture. His emphasis on

rule-following and his painstaking examination of the justificatory role that

reasons play in our lives – that “reasons come to an end somewhere” – is an

original attempt to draw the philosophers’ attention to the normative, social and

irreducibly consensual character of the practice of reason-giving. Contrasting

reason-explanations with causal ones, Wittgenstein elaborates on the

alternative, contextual paradigm he puts forward throughout the Philosophical

Investigations. Ultimately, one is invited to acknowledge the utter vagueness of

the philosophical idea of cause, as well as the diversity of our actual use of this

notion in practice when giving explanations of one’s actions.

Indeed, later developments in the philosophy of mind seem to have

confirmed Wittgenstein’s insights. The invention of the metaphysical notion of

supervenience and the dogma of anomalous monism, despite being offered as

further elaborations of the metaphysics of cause and action, have only produced

more heat than illumination.

The ‘utility’ of the above analysis for IR theory is debatable. Nevertheless,

Wendt (2000) in his reply to Smith (2000) insists that reasons can be causes.

238

This means that consensus over the way one ought to address this philosophical

problem, which underlies the difference between constitutive and causal theory

and as a consequence has implications for both IR theory and empirical

research, is not to be expected anytime soon.

4.4.2 ‘Meta-theoretical hypochondria’: True or an exaggeration?

The perplexed and at times tedious philosophical argumentation which

makes for the arcane164 language which is used when debating SR as an IR

meta-theory brings me to the last issue I will be addressing in this chapter: the

current extensive concern with epistemological, methodological and ontological

issues makes one wonder whether this happens at the expense of dealing with

other more important IR issues which surface in the international political

landscape, or not. Fred Halliday (2000: 247) used Cilfford Geeertz’s term of

‘epistemological hypochondria’ to sharply criticize this penchant for meta-

theorising which characterizes the works of a good number of modern IR

theorists. The term refers to an “obsession with method and with questioning

philosophical issues at the expense of getting on with a job of actually looking

at the world or looking at societies” (ibid).

While Halliday recognizes the importance of having a view on the

epistemologically related issues, he thinks that “to spend the majority of your

time and the majority of your journal space, discussing meta-theory,

164 Reus-Smit defends the validity of meta-theory but not the arcane language in which ‘meta-theoretical bunkerists’, as George Lawson (2007: 775) aptly names them, take refuge in order to make their arguments unintelligible by other IR scholars who work in different areas of interest: “Nowhere here do I defend the cultivated obscurantism that enchants small but diverse pockets of our field, in which ornate language is deployed as a badge of membership more than a medium of communication” (Reus-Smit 2012: 526).

239

epistemology, the critic of ethnocentric assumptions and all the rest of it is a

misuse of time” (ibid).

His critique becomes harder by arguing that such issues lie outside the realm

of IR, not least because none of the questions posed by these issues “can be

worked out solely or mainly through international relations” (Halliday 2000:

247).165

Some years earlier, William Wallace (1996) had launched an attack on the

alleged trend towards over-theorizing in IR, which is presumably being served

by IR theorists who live in the ‘ivory towers’ of the academe (‘monks’). For

Wallace, these academics are enchanted by the fetishism of doing theory for

theory’s sake. This has led them to the nurturing of scholasticism in IR

theorizing. He accuses them of forgetting that “more, and more detailed,

empirical research should be one of first priorities, guided by theoretical

assumptions but intended to inform – and so modify – theoretical assumptions”

(Wallace 1996: 314). As a consequence, Wallace claims, IR has “become too

detached from the world of practice, too fond of theory (and meta-theory) as

opposed to empirical research, too self-indulgent, and in some cases too self-

righteous” (ibid: 316, italics are mine).166 Wallace prefers ‘technocrats’ to

‘monks’; that is, he prefers scholars who, on the one hand, retain a relative

autonomy from the policy universe in order to remain uninfluenced of the

current modes of thinking and are thus able to produce academic work which

could come up to the high standards the IR academic community has itself set

165 He expressed similar views in Halliday (1996: 324). 166 Wallace wrote this paper just before the end of the Third Debate, so his charges for over-theorizing were mainly directed against critical theorists and post-structuralists. However, given that he does not appear to have changed his mind on this issue from the publication of his 1996 paper until now, I think I am entitled to assume that his argument holds even for the current debate on IR meta-theorizing, which, this time, arises from philosophy of science considerations.

240

for the assessment of the works produced under its auspices. These same

scholars, on the other hand, deal with real-world political problems and are

willing to get actively engaged with governments or other policy-making

oriented institutions or relevant policy boards.

Although he may not be aware of the meta-theoretical assumptions of his

own preferred model of empirical inquiry (cf. Reus-Smit 2012: 4), there is a

great deal of truth in Wallace’s claim that IR has moved a long way off from its

point of departure. As Halliday (1996: 318) liked to remind us, “no crisis, no

discipline” and, as E.H. Carr himself stressed, “IR like all academic subjects

emerged as a distinct academic discipline because of a particular crisis in

modern society, in this case of relations between states” (Carr [1961] 2001: 8 -

9). According to E. H. Carr again, “no science deserves the name until it has

acquired sufficient humility not to consider itself omnipotent, and to distinguish

the analysis of what is from aspiration about what should be” (ibid: 9);

theoretical complacency may constitute a great danger to the IR’s future as a

discipline.

This means that in the case of a political science such as IR, it is rather odd,

to say the least, that a great number of papers which are published in IR journals

deal with theoretical issues that are of no relevance to the study of real-world

political phenomena. Indeed, nowadays a large amount of work in academic IR

is not meant to meet the needs and challenges of real world politics, either on

the level of the political practice or on that of theory (something that many

advocates of the current orientations in IR theorizing dangerously

underestimate). Furthermore, it is not only true that very important currents of

thought in IR were born within critical historical circumstances, but also that

241

sometimes these currents of thought have implications for the nature and quality

of international relations that last for long periods (even until today). For

instance, due to “a fine combination of history and theory” (Brown & Ainley

2009: 37), Christopher Hill (1989) has demonstrated that the Second World War

did not bring about “the death of liberal internationalism” and the “triumph of

realism”. What it did bring about is the allowance of “fewer wild swings of the

pendulum between abstract social engineering and cynical power reductionism”

(Hill 1989: 327)

Hill’s suggestion points to a very stimulating way of doing IR Theory with

the ‘assistance’ of History, since it depicts the evolution of political ideas and

theories in accordance to the historical development of real-world politics.

However, coming back to the initial question, one must closely examine

Steve Smith’s reply to Wallace, in which he argues that “agreeing with Wallace

means that academics will run the risk of having to work within the agenda of

the policy community, of being unable to stand back and examine the moral,

ethical and political implications of that choice” (Smith 1997: 515). At the same

time, he also criticizes Wallace’s view of theory as being only explanatory and

problem solving oriented, instead of being concerned “either with understanding

the world or with emancipation” (ibid: 513-14). That given, I would be more

sympathetic with the views Hill has expressed on another occasion on the issues

of the substance of ‘academic IR’ and the challenge posed to it by the “siren

song of policy relevance”, because I find them more balanced and moderate

than the ones expressed by Wallace and Smith on the same issues. This is due to

fact that Hill’s views ascribe to IR scholars a task that strikes a fine balance

242

between avoiding ‘academic snobbery’ while at the same time preserving “a

distinct sense of purpose” (Hill 1994: 19).

Furthermore, with regard to the place of theory in the ‘academic IR’, Hill

claims that “the academic comparative advantage, which applies to International

Relations no less than to any other subject, is a long time-perspective and a

concern with fundamental causation” (ibid: 20) and that “the nature of state, the

causes of war, the problem of rationality in foreign policy, are all subjects which

deserve extensive reflection in their own right and not simply in the margins of

breathless analyses of Maastricht, the Gulf or Boris Yeltsin” (ibid: 21).

Without denying the reasonableness of the converging – although from

different perspectives – views of Halliday and Wallace, one could challenge the

common view shared by many IR scholars according to which the lack of

practical relevance of some large part of IR theory is due to the rapid recent

growth of IR meta-theorizing.

For instance, Christian Reus-Smit claims that

Ιt is a commonplace to bemoan our field’s lack of practical relevance, and to blame

this sorry situation on our penchant for ever-more extensive theorizing over the

analysis of real-world phenomena (Reus-Smit 2012: 525).

However, I think it is an exaggeration to speak of “our field’s lack of practical

relevance.” Ιnstead, I would go for the rather more moderate expression that

there is a lack of practical relevance of some large part of IR Theory, since I

believe one should not ignore that there is a considerable amount of theoretical

work which has nothing to do with philosophy of science considerations and is

243

of substantial practical relevance.167 On the other hand, in order to do justice to

Reus-Smit’s claim above, one should take into account that he bases his

judgment on the observation that “the most comprehensive survey of IR

scholars reports that 85 per cent of respondents believe that there is a gap

between the kind of research we produce and what the policy community finds

useful, and roughly half of these think the gap is widening” (ibid: 526-27).

Furthermore, Reus-Smit argues “first, that the theory versus practical

relevance thesis rests on assertion more than evidence, and second, that it

misconstrues the problem” (ibid: 525), in the sense that

IR’s status…is undermined not by excessive theorizing, but by a series of other

disciplinary handicaps; by IR’s marginal interest in the nature of politics as a

distinctive form of a social action, by the dissipation of the field’s early practical

intent, by the persistent bifurcation of explanatory and normative inquiry, and,

symptomatic of these problems, by the virtual extinction of the figure of

international public intellectual. (ibid: 526)

Although I do not endorse every single point of Reus-Smit’s argument, I feel

aligned with him when he argues that “theory, even abstract meta-theory, can

aid practical knowledge” (ibid: 539). First of all, he is right in arguing that

meta-theoretical reflexivity, along with clarity of purpose, logical coherence,

consideration of alternative arguments, and the provision of good reasons 167 I think that the discussion about SR as a meta-theory of IR is not so important as other ideas and programmes which are already under discussion within substantive IR Theory; for instance, I find much more interesting and maybe fruitful the attempt to bridge constructivism and Realism which has been undertaken by Barkin (2010). Barkin bases his main argument on the crucial and very often bypassed observation that "Realism is a substantive theory of international politics, constructivism is not; constructivists are making a general point about the nature of knowledge in the human sciences, i.e. that it is reflexive and intersubjectively created, and this general point may or may not be compatible with one or other variety of Realism - there is no hostility between the two camps" (Brown 2012: 861-62).

244

(empirical evidence, corroborating arguments, textual interpretations, etc.) are

necessary and sufficient conditions in order for a theoretical work in IR to be

considered worthy of assessment and evaluation. In particular, he indicates that

if epistemological assumptions determine or, at least, affect the questions we

ask, then, if we are conscious of these assumptions, we will also be conscious

of the limitations they pose to the kinds of questions we are ‘allowed’ to ask.

Likewise, if ontological assumptions affect how we see the social universe that

surrounds us and which we are also integral parts of, determining what does

exist within this universe allows us not to neglect to take under consideration

the things that matter. As Reus-Smit says, “the oft-heared refrain that ‘if we

can’t measure it, it doesn’t matter’ is an unfortunate example of epistemology

supervening ontology, something that meta-theoretical reflexivity can help

guard against” (ibid: 533). It is obvious, however, that many IR scholars choose

a theoretical model, or even frame their own, without being aware of the

epistemological, ontological and methodological assumptions which underpin it

and which have implications for the kind of work they can produce within this

certain theoretical and meta-theoretical framework. Moreover, I now come back

to a point I have already made in the second subsection of this chapter,

according to which IR scholars tend to forget that every distinct IR meta-theory

is, as Kurki (2009) points out, ‘politically charged’, in the sense that it has its

own political connotations and implications, which determine (that is, enable or

constrain) the context of the substantive IR theory it underpins. For instance:

Critical realism’s insistence on the language of material reality and causality, for

example, and its suggestion that social life exists as a stratified, layered reality

245

beyond the ‘observable’ could be seen as ‘politically charged’. These

argumentations are widely utilized by critical realists to reject idealist and relativist

strands of argumentation, seen as suggestive of political

conservatism…Simultaneously, the language of layers and depth ontology is used

to reject the atomistic image of society as constituted by autonomous individuals

espoused by many positivist liberals. (Kurki 2009: 445).

The initial question about whether considerations of philosophy of science are

of value to substantive IR theorizing can be reformulated as “Is IR a scientific

field so that it should be grounded on philosophical foundations or not?” Nuno

P. Monteiro and Keven G. Ruby, in their joint paper, “IR and the false promise

of philosophical foundations” (2009a), have first raised this question and

International Theory, the journal in which the above paper was published,

organized a symposium on the topic, “Who needs Philosophy of Science,

anyway?”168

The elaboration on important questions such as the alleged scientific nature

of IR and the feasibility or not of coming up with a single IR foundational meta-

theory, which have been both extensively debated in the framework of the

aforementioned symposium and about which I have discussed in section 4.3.3 of

this chapter, cannot be encountered comprehensively within a single chapter. I

have only provided some indicative arguments on why I am an advocate of the

view that there is no need to search for a foundational meta-theory of IR and I

am all for a ‘pluralist science of IR’, in which differing epistemologies and their

168 Milja Kurki (2009), Patrick Thaddeus Jackson (2009), Fred Chernoff (2009), Raymond Mercado (2009) and James Bohman (2009) have all contributed to this symposium by submitting their critiques of Monteiro’s and Ruby’s paper to the International Theory; the debate has been closed by a reply from Monteiro and Ruby (2009b) but this is certainly not the last word spoken on the question.

246

consequent methodologies bring different aspects of the observed complex

political phenomena within sight.

I would conclude that although the charge of ‘meta-theoretical hypochondria’

advanced against modern IR theorizing has a good deal of truth, one should not

forget that any IR theory relies upon a certain meta-theory and that the

substance of the former is dependent, at least to a certain extent, on the latter’s

assumptions. What matters to every IR scholar is to be conscious of the meta-

theoretical assumptions of his substantive theoretical or empirical work.

At the end of the day, theory and meta-theory are only human intellectual

products, which are characterized by different degrees of abstraction and which

are supposed to help us capture the complexities of political life as that is

deployed on the national and international levels:

Academics quite properly travel the highways and byways of methodology, theory

and history in their efforts to explain political action. But it is the elected

government and its primary institutions that citizens look to in the first and last

instances to secure and improve their lives, and which should be held to account.

When political scientists neglect this fact, and their own role in society, they risk

aridity. (Hill 1991: 247)

Real-world politics are inextricably connected with real people’s daily agonies

for the benefit of their own well-being and future.

4.5 Conclusion

In this chapter I have critically discussed the introduction of SR into IR. I have

primarily focused on the works of the arch-scientific realists Wendt and Wight.

247

I build my argument by taking on board comments on certain aspects of SR,

which were suggested by both proponents, as well as, most crucially, opponents

of SR. I have tried to use a rigorous, analytical approach to the SR

problematique and to avoid generalized expressions about the prons and cons of

the adoption of SR by IR. The conclusions I arrived at are the following:

1) SR was introduced into IR as its proper meta-theory, which is assumed to

underpin substantive IR theories.

2) Unfortunately, this introduction was not accompanied by any examination

of the arguments for and against SR, which have been deployed in the

relevant discussion about both SR and its anti-realist alternatives in the

philosophy of science literature. On the contrary, the standard brand of SR

which is imported and used in IR meta-theory has been considered to be the

definitive and conclusive version of realism. This is mistaken, however,

since the introduction of the standard brand of SR into IR has led to the

neglect of all the shortcomings that characterize SR as it has developed in

the philosophy of science.

3) Furthermore, the miscomprehension of certain concepts and premises of SR

has increased due to the inability of some critics to differentiate between SR

and CR (the latter is a version of SR that relies heavily on Bhaskar’s work,

which, however, is rarely discussed in the SR debate in the philosophy of

science).

4) Wight, but not Wendt, endorses almost the entire corpus of the Bhaskarian

CR and thus becomes a typical critical realist, followed by important IR

theorists, like Patomäki, Joseph, Kurki, Jessop, etc. In IR, if one refers to

SR, one means more often than not CR.

248

5) SR is based on the premise that there is an independently existing world,

which is knowable to us. This world consists of observable and

unobservable entities. The theoretical entities, which are provided by

mature scientific theories, are taken to be true.

6) Following Bhaskar, CR distinguishes philosophical ontology, which

concerns the relationship between the researcher and the world, from

scientific ontology, which concerns the catalogue of things that are taken to

exist and, hence, are available to serve as objects of scholarly research.

7) CR prioritizes ontology over epistemology without committing IR theorists

to a certain set of entities. This leads to an epistemological pluralism and to

a judgmental rationalism.

8) Wendt’s SR is compatible with positivism; moreover, Wendt puts forward

the constitutive explanation, which is complementary to the causal

explanation and constitutes a ‘via media’ and a ‘bridge’ between positivist

and post-positivist methodologies. In contrast, for critical realists, CR is not

compatible with positivism. Wendt’s constitutive explanation rejects the

dichotomy between reasons and causes, in that one can accept that reasons

can be causes. For this reason, the reasons vs. causes debate is of central

importance to Wendt’s SR.

9) Wight argues that the explaining versus understanding debate belongs to the

realm of methodology rather than that of epistemology. Furthermore, he

argues that if something in this debate is of value to his own theory is that

the distinction between these two terms is based on ontological

considerations about the social world. Therefore, the reasons vs. causes

249

debate, which arises from the explaining vs. understanding debate, is not

crucial to CR.

10) Wendt has used SR in order to introduce the notion of unobservable entities

in the IR vocabulary and thus legitimize the theorization of unobservable

structures in the study of the agent-structure issue within IR theory. His aim

was to show that neither the alleged individualist ontology of neorealism

nor the structuralist ontology of the world-system theory is adequate to

explain the formation of international politics. He thinks that the

explanatory scheme of the agent-structure relationship can help one explain

not only how international politics are shaped and reproduced but also how

these can change.

11) In contrast, Wight takes agents and structures to be ontologically strictly

independent, providing us with certain typologies of structures and agencies

which, however, are not always compatible with one another.

12) Wendt’s STIP is an important piece of theoretical work. This theoretical

gravity is mainly found in its second part, which concerns his social theory

of international politics and in which he deploys his own constructivist

theory. In it Wendt emphasizes the role of ideas but also acknowledges the

limits, which are imposed on them by ‘rump materialism’. Per contra,

Wight’s intention was not to provide a substantive theory of IR but, instead,

to argue that IR should become a science without need to resort to

positivism.

13) All the above have been contested by eminent IR scholars who have

criticized SR and CR on two fronts: first, in terms of their inner logic,

250

consistency, coherence and comprehensiveness as theories and, second, in

terms of their ability to help the building of substantive IR theories.

14) Wendt’s ‘rump materialism’ is contested because, as Steve Smith, among

others, has argued, Wendt’s ambiguous statements about the relationship

between the material and the ideational in his STIP allow one to doubt the

status of his ‘rump materialism’. I argued that, although Wendt’s

argumentation is not free from inconsistencies and contradictions, his thesis

is not ‘ideas all the way down’ but rather ‘ideas almost all the way down’;

otherwise his constitutive explanation as a via media methodology makes

no sense.

15) However, Smith’s further point that in both of the above cases the

ideational and the material are very different kinds of stuff, and this seems

to make naturalism impossible, is at least debatable.

16) According to the majority of their critics, Wendt and critical realists have

not succeeded in coming up with certain empirical research proposals and

this is considered to be the major shortcoming of both SR and CR.

17) Critical realists and Wendt believe that meta-theoretical choices are

‘politically charged’ and, as such, have implications for the choice of the

substantive IR theory. This is quite a plausible position to hold.

18) More generally, however, the consideration of meta-theoretical issues is of

some value, since every IR scholar should be aware of the meta-theoretical

premises he uses in order to underpin his own work. This, nevertheless,

must not lead to a ‘meta-theoretical hypochondria.’

19) If one takes under consideration the deadlocks one reaches when one

discusses the meta-theoretical issues which are posed by SR and CR, it is

251

legitimate to conclude that further intensive involvement in this kind of

discussions does not merit any progress in IR theorizing.

20) Chris Brown has argued that CR can do the IR discipline service only if it

leads to a resurgence of interest in genuine Marxist studies in IR and, for

the sake of pluralism of ideas in IR theory, I absolutely agree with that

position.

21) I strongly support a ‘pluralist science of IR,’ which must be more inclusive

even than Jackson’s, in the sense that it must incorporate theories of

language turn and speech act theories (as Onuf suggests). The above

theories must be included in the ‘methodological apparatus’ of IR, even if

this implies that, in this case, SR would not be justified branding itself as a

science.

22) To pay lip service to the notion of science implies that IR should endorse as

legitimate epistemologies all the post-positivist ones, not that positivism

should be abandoned. Therefore, in the above broad definition of the

‘science’ of IR, or whatever else one would like to call it, positivism should

retain its important – however not overarching – role. Otherwise,

quantitative and formal methods would have to be abandoned along with

the questions that seemed to call for these very methods.

Overall, I conclude that the introduction of SR/CR into IR does not stand up to

the scientific and critical realists’ expectations. Consequently, one may wonder

if the study of this issue deserves the intense intellectual labour of many leading

IR theorists, proponents and opponents of SR alike. Sometimes, however, the

journey towards knowledge might yield more benefit than the answers at the

252

final destination of an intellectual endeavour. Such seems to be the situation

with the meta-theoretical debates within IR.

5

ALEXANDER WENDT’S QUANTUM SOCIAL SCIENCE

AND INTERNATIONAL RELATIONS

5.1 Introduction

Despite the fact that the use of SR in the Philosophy of Science and IR has been

extensively discussed in the previous three chapters of the thesis, one more

issue should be added to my examination. This issue is Wendt’s quantum social

science project and its implications for IR Theory, and emerged when

Alexander Wendt (2006a) devoted a significant portion of his reply to the

critics of his celebrated STIP to a provocative outline and defence of a quantum

approach to the social sciences. In this venture he aspires to initiate a new

science of IR, as part of a new social science as a whole. Thus Wendt grounds

his newly developed theoretical investigations in quantum mechanics, while

underpinning them by using a revised SR as their meta-theory.

The aspirations raised by his new research program are well summarized by

Griffiths, Roach and Solomon when they write that: “Wendt has drawn on

quantum theory to probe the limits of international theory” (2009: 158).

Furthermore, they argue that Wendt’s “primary aim is to suggest the

possibilities of what he calls ‘capacity for collective self-consciousness’” (ibid:

158-159), meaning that a quantum social science may offer more sophisticated

and contextually enriched methods for studying the formation of ‘collective

254

self-consciousness’ than the ones we already have at our disposal. They also

believe that Wendt has been attracted to quantum theory by its “non-

reductionist and non-deterministic” character. The latter led him to envisage a

‘quantum social science’, in which quantum theory “has to be applied in a

systematic manner to world politics” (ibid: 159). As for the epistemological

goals of this undertaking, they claim that “in his view, such theory holds

important implications for developing a new non-foundational epistemology,

which might help to further reconcile critical theory with mainstream

approaches” (ibid: 159).

A critical step-by-step presentation of Wendt’s argument will help clarify its

structure and content as well as its shortcomings. Wendt’s starting point is that

the mind-body problem – in particular, the account of consciousness and

intentionality – is a fundamental problem for social science that has not, as yet,

found any satisfactory solution in the philosophical or scientific literature

(Wendt 2006a: 185-189). This prompts him to put forward what he

acknowledges to be a ‘heretical thought’: the limitations of both contemporary

philosophy of mind and social science stem from the common underlying

assumption that the connection between mind (‘ideas’) and body (‘matter’)

must be informed by the world-view of classical physics (Wendt 2006a: 183).

Thus he advances two claims: First, that the quantum consciousness hypothesis

– namely, that “consciousness is a macroscopic quantum mechanical

phenomenon” – is true. Second, that the quantum nature of consciousness does

have implications for social science (Wendt 2006a: 183-184).

Wendt then goes on to devote the rest of his exploration of the quantum

underpinnings of social science on the following tasks: First, he presents a non-

255

technical review of the conceptual innovations of quantum theory (Wendt

2006a: 190-193), then he presents the various aspects of the quantum

consciousness hypotheses in connection with a variant of panpsychism (Wendt

2006a: 193-196) and he gives an outline of a quantum model of man that

allegedly accounts for both teleology in human action and freedom of the will

(Wendt 2006a: 197-199). Finally, he gives a sketch of a quantum model of

social systems within which one is supposed to get a glimpse of the collective

unconscious, the collective consciousness and human interaction (Wendt

2006a: 200-205).

Wendt is clearly aware that his quantum research program in social science

is as complex as it is risky. As a consequence, he is content with merely

offering an outline of how a possible course of argument for a quantum social

science might be formulated. Accordingly, it may appear premature to embark

on a detailed assessment of Wendt’s quantum turn in focus. Indeed, Wendt does

not offer detailed arguments, explanations or testable hypotheses.

Consequently, one might regard Wendt's whole approach as a hodge-podge of

contentious speculations, the one stacked onto the other. Nevertheless, it is not

premature to identify some crucial methodological, scientific, and philosophical

shortfalls in the ‘quantum Wendt’, as this will serve as a cautionary alert to any

similar intellectual endeavours.

This is precisely the aim of the chapter at hand. Section 2 identifies a

number of weaknesses in Wendt’s methodology. Section 3 focuses on physics

in an attempt to expose Wendt’s misconceived basic concepts of quantum

theory and his misconstruing of this theory’s ‘functioning’ in the real physical

world. Section 4 argues that Wendt is not explicit in his approach as to the

256

metaphysical principles he uses in order to underpin his program. These three

sections (2 to 4) constitute a fairly comprehensive investigation with regards to

the methodological, purely scientific and metaphysical weaknesses of Wendt’s

quantum program. In section 5 I will draw readers' attention to the implications

of these results for the way Wendt grounds his ‘new IR science’ on quantum

mechanics. Wendt’s new explorations in IR Theory will be exclusively dealt

with for this reason. I intend, inter alia, to examine Wendt’s idea of the

international system as a hologram along with a consequent new approach to

the ‘level of analysis’ problem based on the former. Both have been recently

presented in his piece, “Flatland: Quantum Mind and Social Science” (Wendt

2010),169 a sequel to the “Social Theory as Cartesian Science – An auto-critique

from a quantum perspective” (Wendt 2006a). It is in this context where lies my

undertaking of proving that Wendt’s latest theoretical achievements do not

necessarily presume abandonment, but only a revision of scientific realism as

the meta-theoretical framework of his differing versions of STIP.170 The last

section summarizes the conclusions that can be drawn about the prospects of

such a tentatively sketched program. To sum up, what is at stake in this chapter

is to discuss Wendt’s ‘quantum physics’ and its extensions to social science and

the theory of International Relations (IR). The fact that ‘quantum Wendt’ has

not attracted much attention from IR scholars171 so far may attribute an added

value to the aforementioned venture.

169Wendt in his personal webpage (http://mershoncenter.osu.edu/expertise/spotlight/Wendt.htm) informs us that this piece is the basis for his current project, Quantum Mind and Social Science. A preliminary outline of this project is to be found in Wendt (2006b).170 Wagner and Gebauer (2008), when referring to Wendt (2006a), talk of a “revised STIP” or even a “STIP 2”.171 Wagner and Gebauer (2008) make this point at the beginning of their paper (“Judging by the marked absence of responses to Alexander Wendt’s Auto-Critique from a quantum perspective thus far, it would seem that the academic community have either gone back to their books and are still there, that it would prefer to ignore Wendt’s paper as irrelevant to the academic debate,

257

5.2 Methodological issues

Wendt does not support his claims that facts concerning human consciousness

and human interaction are explainable by appeal to quantum theory via the use

of precise and detailed arguments. Rather, he postulates that human beings and

social systems exhibit quantum features (e.g. are described by wave functions

subject to collapse) and tries to assess the explanatory gains from such

postulates. The assessment relies on weak analogical arguments in the context

of which aspects of human and social life are compared to aspects of the

behaviour of quantum matter. For instance, the transition from an unconscious

state to a conscious one is seen as similar to wave-function collapse and the

sharing of meanings and knowledge among members of a society is seen as

similar to quantum entanglement. The understanding of the mentioned aspects

of quantum systems is, in turn, supported by arguments from authority: Wendt

cites, here and there, views of theorists he takes to be specialists on quantum

physics and its philosophical implications. Moreover, Wendt has recourse to

consistency considerations only on occasion. He claims, for instance, that one

reason to take panpsychism seriously is that “it is consistent with quantum

theory” (Wendt 2006a: 195). The methodological pitfalls with such stratagems

are both grave and multiple.

To begin with the most obvious, mere consistency does not necessarily entail

evidence of truth. What’s more, a hypothesis like panpsychism, i.e. “the view

that something like human consciousness goes all the way down to the sub-

or that is simply waiting to see what will happen”) and repeat it in their conclusions, as well (“This article began by emphasizing the marked lack of responses Wendt’s Auto-Critique has drawn so far. Oliver Kessler’s article in 2007 aside, there has been no meaningful engagement with his modified STIP, a deficit that we hope to address in part here”). However, they do not mention the other IR scholar who has seriously embarked on a critical appreciation of the ‘quantum Wendt’, namely Keeley (2007).

258

atomic level” (Wendt 2006a: 195) is trivially consistent with any (consistent)

physical theory, simply because such a theory does not even talk about

consciousness. The ‘venerable’ history of panpsychism in Western philosophy,

to which Wendt himself alludes, offers just another indication of this point.

The issue that characterizes more pervasively Wendt’s quantum approach to

social science, however, is the deployment and use of analogical arguments.

Here, again, the methodological fact to be observed is rather simple: similarities

alone do not pave the way to explanation. Suppose some entity X is similar to

another entity Y with respect to some feature R and suppose, further, we do

have an explanation of why Y exhibits R. All this need not necessarily imply

that we have an explanation of why X exhibits R, since X and Y may be wildly

different entities that simply ‘happen’ to exhibit a formal similarity.172 In

addition, the evaluation of any argument which is based on an analogical

methodology must take into account all known relevant similarities and

dissimilarities.173 Lastly, it is widely accepted in the philosophy of science that

genuine explanation cannot be exhausted by a mere reduction to the familiar

(see, e.g., Salmon 1992: 14).

Wendt’s quantum program does not fare much better either for in this case

one cannot even talk about ‘reduction to the familiar’ since quantum theory is

172 The early hydrodynamic interpretations of quantum mechanics present a clear example. The interpretations were based on structural similarities between the equations of Schrödinger’s wave mechanics and those governing the flow of a nonviscous irrotational fluid subject to conservative forces (e.g. the equation of continuity). Given that the former equations were supposed to govern atomic physics whereas the latter were applicable to the idealization of a continuous fluid, these interpretations appeared to attempt to account for the behaviour of atoms by appealing to theories that deliberately disregarded atomicity. See, Jammer (1974: 33-38). 173 This follows from the requirement of total evidence for the applications of inductive logic (see Salmon 2007: chapter 4). Wendt seems to overlook this requirement systematically. For example, as Keeley (2007: 427) persuasively argues, there are significant dissimilarities between human decision processes and wave-function collapse: “Sometimes apparent decisions are nowhere near this decisive. Instead, competing strands of thought continue even into the implementation process, while the decision itself represents a botched compromise by contending parties rather than some definite resolution”.

259

notorious for its interpretive puzzles. We do not, as yet, satisfactorily

understand the workings of the quantum world and, accordingly, it is extremely

doubtful whether an appeal to such a shaky understanding may shed light on

queries regarding consciousness, society, international relations, or other social

science puzzles. Wendt is, of course, aware of this problem. He writes:

However, what the quantum is, nobody knows. Which is to say, even though

physicists know how to use quantum theory, they do not understand it, what it is

telling us about the nature of reality. Quantum metaphysics is ‘under-determined’

by its physics… and as such requires an ‘interpretation’ of the theory. Philosophers

have been intensely debating which interpretation is correct since the 1930s, and

show no signs of stopping soon. Over a dozen interpretations now exist, with

metaphysics that are not only wildly different, but simply wild. (Wendt 2006a:

189-190)

Nevertheless, Wendt still fails to realize the methodological strictures one has

to adopt in the face of this situation. Indeed, in undertaking a project like

Wendt’s, one has to be clear about both the quantum theory and the

philosophical interpretation of that theory one relies upon. On the contrary,

Wendt remains eclectic and ultimately unclear on both physical and

philosophical grounds (see Sections 3 and 4 below).

Arguments by appeal to authority pose another issue. Such arguments are

persuasive only if the authority cited is really an authority on the matter under

consideration and there is substantial agreement among authorities on the

matter at hand (see Salmon 2007: 118-120). But Wendt appears to violate, at

places, each of these conditions. Most serious researchers on quantum theory or

260

the philosophy of quantum theory would not recommend Zukav’s The Dancing

Wu Li Masters (Zukav 1979), even as good popular book in which the

experimental findings of quantum theory are clearly described (compare Wendt

2006a: 191 and endnote 19).174 On the other hand, Wendt, whether intentionally

or not, regularly brushes aside disagreements among experts and covers up

differences between physical concepts. To illustrate this point, consider the

following excerpt about consciousness:

Although there is disagreement among quantum consciousness theorists about

where precisely consciousness is ‘located’ in human beings, in my view the most

plausible answer is in the collapse of our wave functions. This process happens

continually as we interact with the environment, providing a basis for our

experience of a stream of consciousness…. In other words, the desires and beliefs

which the rationalist model of man sees as causing behaviour actually do not exist

until behaviour takes place – before that point the Self is a superposition of

multiple and mutually incompatible desires and beliefs. (Wendt 2006a: 197-198)

In the excerpt above, Wendt begins by acknowledging disagreement among

quantum consciousness theorists and then sets aside the disagreement in order

to put forward his own conjecture. In the very next sentence, he vaguely

associates two distinct physical concepts – namely, ‘wave-function collapse’

174 Moreover, there is no final agreement on whether one can comprehend the conceptual issues of quantum theory without immersing oneself into physical and mathematical niceties. For instance, Albert’s Quantum Mechanics and Experience (1992) is an authoritative and illuminating presentation of these issues with minimal (‘high-school’) requirements as to the reader’s background knowledge in physics and mathematics.

261

and ‘interaction with the ‘environment’,175 before he finally goes on to draw

some conclusions regarding human decision-making and behaviour.

Due to the aforementioned methodological aspects, Wendt’s quantum

conjectures can, at best, be regarded as metaphors. It is worth noticing here that

this kind of work is not rare in the literature on quantum consciousness.

Atmanspacher assures us that it is rather common:

There are quite a number of accounts discussing quantum theory in relation to

consciousness that adopt basic ideas of quantum theory in a purely metaphorical

manner. Quantum theoretical terms such as entanglement, superposition, collapse,

complementarity, and others are used without specific reference to how they are

defined precisely and how they are applicable to specific situations. For instance,

conscious acts are just postulated to be interpretable somehow analogously to

physical acts of measurement, or correlations in psychological systems are just

postulated to be interpretable somehow analogously to physical entanglement.

Such accounts may provide fascinating science fiction, and they may even be

important to inspire nuclei of ideas to be worked out in detail. But unless such

detailed work leads beyond vague metaphors and analogies, they do not yet

represent scientific progress. (Atmanspacher 2006: 8)

The excerpt dissects accurately the profile of the work Wendt has produced so

far on quantum approaches to the social sciences.

What is striking in Wendt’s case, however, is that his aim is not to provide

one more metaphor as to the adaptation and use of both quantum mechanical

175 Here and elsewhere Wendt seems to confound quantum state-vector reduction with decoherence.

262

logic and concepts within the theoretical framework of social science and IR

theory, as one would eventually have preferred to consider it.

The above is actually the goal of the venture on which Dimitrios Akrivoulis

-not Wendt- has embarked in his own PhD dissertation, The quantum politics

metaphor in International Relations - Revising American Newtonianism

(Akrivoulis 2002), which, according to Ben Wagner and Jonas Gebauer,

constitutes “the most complete metaphorical consideration of quantum politics

to date”, for its author “undertakes a long and detailed study of a paradigmatic

quantum shift in the political science” (Wagner and Gebauer 2008: 9-10).

Akrivoulis, influenced by Rob Walker’s writings in IR Theory (cf. Walker 1991

and Walker 1993), also takes on board Ricoeur’s seminal philosophical work in

hermeneutics. Thus, considering imagination as a dimension of language he

differentiates between the social functions of the Newtonian imaginary (which

the ‘established’ social science is presumably based on)176 and the social

functions of the quantum imaginary. He does all this, in order to conclude,

among others, that

It is here that the fundamental ambiguities of a quantum imaginary could be placed

beside, juxtaposed with and contrasted to the ones of the Newtonian imaginary.

And here the regressive analysis of meaning that we follow, when examining the

positive and negative traits of the Newtonian ideology (integration, legitimation,

distortion), could also prove helpful in the exploration of the respective traits of a

quantum utopia (challenge, possibility, and escapism). Given the above, it is

176 Τhere are, on the other hand, those who claim that “for conceptual and empirical reasons” the quest for predictive theory in IR “rests on a mistaken analogy between physical and social phenomena”, whereas “evolutionary biology is a more productive analogy for social science”. In this vein, they have developed ‘hard’ and ‘soft’ versions of this analogy, too. For such an approach, see Bernstein, Lebow, Stein and Weber (2000).

263

perhaps not difficult to access that, first, whereas the Newtonian imaginary

functioned in an integrating manner establishing a kind of social bond and

reaffirming a sociopolitical order that was taken as natural and given, a quantum

imaginary would aim at denaturalizing this order by subverting or challenging the

given forms of social bond and political relating. (Akrivoulis 2007: 10)

On the other hand, in his attempt “to use quantum mechanics as the foundation

of a theory of international politics” (Wagner and Gebauer 2008: 10), Wendt

has gone far beyond a mere argument in favour of a new quantum metaphor.

There is no clearer evidence of this claim’s truth than the following:

If the quantum consciousness hypothesis is true then the elementary units of social

life, human subjects, are quantum systems – not just metaphorically or by analogy,

but really. (Wendt 2006a: 196)

Or, even more precisely,

This book project explores the implications for social science of thinking about

human beings and society as quantum mechanical phenomena. In the past there has

been very limited discussion of this question, but only as an intriguing analogy and

thus it had essentially no impact. My suggestion is that man (sic) and society really

are phenomena. (Wendt 2006b)

While a metaphor is usually based on assumptions of the sort “as if”, here

Wendt appears to neglect the metaphorical or analogical use of the core

elements of his theory, by taking them to be real.

264

The implications of Wendt’s idiosyncratic –to say the least- interpretation of

quantum mechanics for the study of the philosophical foundations of social

science and IR, make him the sole representative of ‘strong quantum politics’,

whereas Akrivoulis, along with others, are defenders of what Wagner and

Gebauer (2008: 11) tend to call ‘weak quantum mechanics’. The scholars

belonging to the latter category make use of quantum mechanics in a

metaphorical manner when studying international politics, while Wendt calls

for a ‘new social science’, a ‘new IR Theory’ based literally on quantum

mechanics.

Arguably, it was these expectations raised by Wendt’s ‘strong quantum

politics’ for radically shifting and reorienting modern IR Theory that led Barry

Buzan to call his colleagues to arms by advising them to run back to their books

and study not only IR Theory and Philosophy of Science but Quantum

Mechanics as well, in order for them to be able to come to grips with this new

intellectual challenge presented by Wendt’s latest work:

The good general Wendt finds here an attack that should disorientate his enemies

on both fronts, and force the more energetic and creative of them back to their

books for some overdue cross-disciplinary re-education. This is ‘new International

Relations’ with vengeance. (Buzan 2006: xvi)

More recently, Simon Curtis and Marjo Koivisto (2010) – in the context of their

attempt to reformulate the relationship between Science and History in

International Theory with the help of the recent alleged transcendence of the

assumption of incommensurability between scientific and historical frames of

inquiry – make a reference to the role positivism has had and still aspires to

265

maintain as the basic methodology for both natural and social sciences. In the

context of the above inquiry, they conclude that Wendt’s quantum social

science to be a rather promising new positivist alternative:

The early positivists thus had faith in the unity of science: a single scientific

method to gain access to the secrets of both the natural and social worlds. … In

their view, human beings were physical objects located in physical space, and, just

like other physical objects, they should theoretically be able to be integrated into

the physical sciences, with their behaviour deducible from their material needs.

This would allow the predictive capacity (and, in the case of historical knowledge,

retroductive capacity), necessary for a ‘social physics’. The notion that the

activities of human beings can be predicted by reference to their physical needs is

an essential element of mechanical approaches to formulating behaviourist science.

An interesting recent move that has clear similarities with this tradition is

Alexander Wendt’s fascination with a quantum social theory of international

politics. (Curtis & Koivisto 2010: 438-439)

However, although I appreciate Buzan’s pioneering work in IR Theory, which

made him an authority in the field and whose claims about IR theory must

always be carefully considered, and despite the fact that I do not underestimate

Curtis’ and Koivisto’s rather positive stance vis-à-vis Wendt’s latest intellectual

endeavour, I still remain, at the very least, sceptical in terms of the potential

that Wendt’s new approach to social sciences has, if it is to be regarded as

something more substantial than a mere metaphor.

In order to adduce further support for this claim, I will now turn from

methodological to more ‘substantive’ theoretical matters.

266

5.3 Wendt's ‘Understanding’ of Quantum Physics

As already said, the main thesis in Wendt’s paper is that the current approach to

Social Sciences (SS) and World Politics (WP) is inadequate and that a radically

new mental framework is needed. The latter must be based, literally, on

Quantum Physics (QP). That means that consciousness results from the brain

being a macroscopic quantum system in a coherent state (at least partly); and

treating it as such might shed light in the study of SS and WP.

As the author recognizes explicitly, this thesis is highly speculative, to say

the least, in both of its two assumptions: (a) that consciousness is a

macroscopic quantum mechanical phenomenon, and (b) that this may have

implications for SS and WP.

The arguments advanced by Wendt in support of the second assumption can

be reduced to the general statement that the inner workings of the human brain

(how the brain becomes mind) influence the way we perceive reality and, as a

result, influence how the scientific truths are reached by the collective mental

processes of the scientific communities. Such a statement is too general to be

persuasive. Although it is recognized as a deep fundamental question, it cannot

receive a scientific answer at least until the consciousness formation (or, more

generally, the brain/mind question) is resolved. Until then, it cannot get a

scientific answer even in connection with much simpler (than the study of SS

and WP) frameworks, such as the relation between mathematics and physics or

our inability to perceive space with more than three dimensions.

I conclude my critique of the assumption (b) in Wendt’s thesis by saying

that it is way too premature to argue in favour of the influence of the mind-

267

body problem on the way we study SS and WP. Moreover, I will return to this

issue in section 4 as well, where I will examine the metaphysics of the mind-

body problem in more detail.

Before turning to the arguments that Wendt puts forward in favour of his

first assumption (that consciousness is a macroscopic quantum mechanical

phenomenon), it is worth pointing out that I will enter the realm of Physics,

where concepts have precise mathematical definitions and statements ought to

be subject to falsification by quantitative observations/experiments by the

flawless operation of existing technology.

5.3.1 Aspects of Quantum Physics Relevant for Wendt’s Theory

Wendt starts by stating that the laws of physics impose constraints on SS; this

is, of course, true for any science of higher complexity than physics (chemistry,

biochemistry, biology, ecology, SS). Then he asserts that the laws of physics

can be either classical or quantum and that the former is inadequate for

studying the brain/mind problem, because classical mechanics is too materialist

to capture the deep essence of consciousness.

As I will point out further on, the laws of physics are quantum mechanical,

not classical; period.177 Classical mechanics becomes simply an adequate

approximation to quantum mechanics only under certain limiting

circumstances. The structure and the properties of matter at all scales, from the

protons and neutrons to atoms, molecules, living cells, ordinary macroscopic

solids and liquids, planets, stars, white dwarfs, etc., are what they are because

they obey quantum mechanical laws. The universal physical constant ħ, the

177 The analysis which follows is based on Feynman, Leighton and Sands ([1965] 1989), Griffiths (2002) and Paul (2008).

268

‘trade mark’ of quantum mechanics, is a determining factor in the structure and

properties of all forms of matter. The density, the compressibility, the thermal

conductivity, etc., of the simplest macroscopic system, e.g., of a piece of metal,

such as a copper wire, are what they are because of quantum mechanics; their

values involve the universal quantum constant ħ. If classical mechanics were

true, matter, living or not, would not exist; the whole world would eventually

collapse to one or more black hole-like entities.

The widely accepted statement, apparently adopted by Wendt, that the

microscopic world is the realm of quantum mechanics, while the macroscopic

one is the realm of classical mechanics, is misleading. The correct statement is

that classical mechanics fails completely to account for microscopic motions,

while it is a fully adequate approximation for macroscopic motions only (not

properties) of both macroscopic and microscopic entities. Keep in mind though

that as a result of the atomic structure of matter, microscopic motions

determine the structure, the properties and the functioning, if any, of

macroscopic objects. This is the reason the world is at all scales quantum

mechanical. Consequently, the statement that the human brain functions by

obeying the laws of quantum mechanics is as obvious to a research physicist as

the statement that the properties of a piece of metal are determined by quantum

mechanical laws. What is not trivial is the highly speculative and intriguing

idea that parts of the brain are in a coherent quantum state and, if this is so, it

may have implications for the brain/mind problem.

One may wonder what a coherent quantum state is. To provide any

reasonable (no matter how imprecise as a result of not using the mathematical

language) answer to this question, one needs to present a few basic aspects of

269

the quantum theory. To this end, I will follow Wendt’s suggestion to introduce

quantum mechanics by contrasting it to classical mechanics, and then I will

present my critique of Wendt’s views.178

In classical physics two fundamental entities are introduced: elementary

particles and waves. Elementary particles are local point-like objects,

indivisible and discrete (in the sense that their number in any system is

obviously a non-negative integer). Particles in their motion follow a trajectory

(or an orbit, if you prefer) which is a mathematical line (not necessarily a

straight one); by knowing the position and the velocity at any point and the

total force acting on the particle, then its trajectory is uniquely determined.

In contrast, waves, which exist in a medium (continuous or discrete) and

even in vacuum, are non-local (in the sense that they have a non-zero extent in

space) and are of continuous nature (a macroscopic classical wave does not

consist of a non-negative number of elementary constituents); their motion is

governed by partial differential equations. Finally, the waves exhibit the

characteristic wavy feature of interference (the waves starting from different

sources may cancel each other at the same regions of space, while elsewhere

they may overstrengthen their sum). The phenomenon of wave interference is

easily visible by throwing two stones in the quiet waters of a lake and watching

what happens as the two concentric ripples meet each other. However, if we

throw a large number of stones randomly, the resulting waves are so

complicated and disordered that the interference phenomenon is not visible

anymore. For the same reason we cannot see the interference in the light

coming out from two different ordinary lamps; the waves arriving from these 178 However, for a clearer and more detailed introduction to the basics of quantum mechanics see Feynman’s description of the two slit Gedankenexperiment in “Quantum Behavior”, which can be found in Feynman , Leighton and Sands ([1965] 1989: 1-1 – 1-11).

270

two lamps are disordered superpositions of many uncorrelated waves, so that

the interference cannot become visible. In contrast, if we split the beam from a

laser into two sub-beams, the interference of light waves becomes clearly

visible as alternating dark and bright regions on the screen. If each of two

waves is the superposition of many ‘ordered’ and ‘correlated’ waves so that the

interference phenomenon is observable, we call each wave coherent; if each of

the two waves is a ‘disordered’ and ‘uncorrelated’ superposition of waves as to

make the interference unobservable, is called decoherent. A coherent wave can

become decoherent through interactions with the environment; we say then that

the wave ‘decoheres’.

To summarize: According to classical mechanics, elementary particles are

local, indivisible, discrete entities, following a trajectory in their motion; waves

are non-local, continuous entities exhibiting the phenomenon of interference.

However, all experimental facts show that:

(a) Actual elementary particles (i.e., the ones existing in Nature) are point-

like indivisible, discrete entities which, however, do not follow a trajectory in

their motion but they move as waves exhibiting the phenomenon of

interference. Thus, actual particles have some properties of classical particles

and some properties of classical waves. This is the reason they are named

wave-particles.

(b) Actual waves (i.e., the ones existing in Nature) move as classical waves

exhibiting the phenomenon of interference, but they consist of elementary

indivisible, discrete units. Thus, the actual waves have some properties of

classical waves and some properties of classical particles. It is only reasonable

to name them particle-waves.

271

The conclusion is the so-called wave-particle duality, one of the two

cornerstones of modern science (the other is the atomic structure of matter).

The main task of physics, following these experimental facts, was to

construct a coherent (mathematical) formalism, which will merge together

particle and wave features that seem mutually exclusive. Quantum mechanics

achieves just that but at a price:

(a) Instead of the familiar concept of trajectory, a wave function is

employed for the description of the motion.

(b) The physical information extracted from the wave function is of a

probabilistic nature (although 100% probability, i.e. certainty, is not unusual).

E.g., if you measure the orientation of the spin of the electron, quantum

mechanics may predict that the result will be spin-up with probability, let us

say, 70%, and spin-down with probability 30%. Before making the

measurement, both eventualities (70% up, 30% down) are present; if the result

of the measurement is, e.g., spin-down, from this point on the electron will be

in a state with spin-down and the next measurement of its spin will be spin-

down with probability 100% (i.e., with certainty).

(c) This feature is the third part of the price paid by merging together

seemingly exclusive particle and wave properties. The measurement affects the

measured system by forcing it to be immediately after the measurement in a

state consistent with the results of the measurement. This is known as the

wave-function collapse (the term collapse is more appropriate for measuring

the position; here, however, we are using the term for any measurement).

Probably the most striking and counterintuitive aspect of the role of

measurement in QM appears in entangled systems. Two systems A and B are

272

said to be entangled if their quantum states are correlated - that is, not

separable (see Economou 2012). A famous example of entangled systems was

considered by Einstein et al. in the renowned ‘EPR paper’ (Einstein, Podolsky

& Rosen 1935), in which a particle decays into two photons moving in

opposite directions and having opposite spins, meaning that A is up while B is

down and A is down while B is up. The result of measuring the spin of B will

be either up or down with a probability of 50%. However, if the measurement

of B is done immediately after A has been found to be down, then the result

will be up with certainty, in spite of A and B being separated by arbitrary large

distance. This ‘spooky action at a distance’ led Einstein to question the

completeness of quantum theory as representing the physical reality. 179

Needless to say that the Gedankenexperiment proposed by EPR was actually

179 For instance, a very recent and representative example of the ongoing discussion on these issues comes from a paper written by Pussey, Barrett & Rudolph (2012). Its title reads “On the reality of the quantum state”. Let us see how they summarize their main argument in the beginning of this publication:

Quantum states are the key mathematical objects in quantum theory. It is therefore surprising that physicists have been unable to agree on what a quantum state truly represents. One possibility is that a pure quantum state corresponds directly to reality. However, there is a long history of suggestions that a quantum state (even a pure state) represents only knowledge or information about some aspect of reality. Here we show that any model in which a quantum state represents mere information about an underlying physical state of the system, and in which systems that are prepared independently have independent physical states, must make predictions that contradict those of quantum theory (Pussey, Barrett & Rudolph 2012: 475).

In other words, this work is based on two assumptions. The first assumption is that there is a ‘real physical state’ of a system, which is objective and independent of the observer. The second is that systems prepared independently have independent real physical states. These rather obvious assumptions lead to the following no-go theorem: If the quantum state (ψ) merely represents information about the real physical state of a system, then the experimental predictions obtained contradict those of quantum theory. Therefore, the authors conclude that the wave function (ψ) is real in the sense that it does not represent mere information about an underlying physical state of the system, but is the physical state itself of the system.

The philosophical implications of this paper for scientific realism (SR) are inextricable to the extent that the central conclusion of the paper concerns the very nature of the mathematical skeleton of a fundamental physical theory, such as quantum mechanics. As has been shown in the second chapter of this thesis, this issue belongs to the core of the problematique of SR in the philosophy of science.

273

performed in 1982 by Aspect et al. (Aspect, Grangier & Roger 1982) and the

resulting data were in full agreement with the predictions of QM.

Let me now show why wave-particle duality is of fundamental importance

in accounting for the world structure (in the framework of the atomic idea).

According to the latter, everything is made up of indivisible, microscopic

wave-particles, which attract each other through the action of interactions

(which are made of particle-waves). From the level of atomic nuclei all the way

to the level of an asteroid there are three types of wave-particles: protons,

neutrons (organized into nuclei through the action of the strong interactions)

and electrons (trapped around the various nuclei by the action of the

electromagnetic interactions to form electrically neutral atoms). Atoms make

molecules, ordinary solid and liquid matter, living organism, etc. However, the

interactions, as a result of their overall attractive nature, would squeeze all

these structures to one or more black-hole-like entities, if left unopposed. What

opposes the interactions and prevents their catastrophic squeezing, making,

thus, the existence of matter inside and outside us possible? It is just the

wave/particle duality! This duality leads to Heisenberg’s uncertainty principle,

which in turn implies that α wave-particle trapped within a finite volume

acquires necessarily non-zero kinetic energy; the latter cannot be smaller than a

value proportional to ħ2 and inversely proportional to the product of the mass

of the wave-particle and the volume to the power of two thirds.

The final fundamental conclusion is that the various structures of matter

exist, because the squeezing pressure of the interaction is counterbalanced by

the expansive pressure of the quantum kinetic energy. This picture of the world

can be easily tested quantitatively by comparing the values it produces for

274

various quantities (such as the size of atoms, the length of chemical bonds, the

density of various solids, etc.) against the corresponding experimental values;

notice that, if we set ħ=0, namely if we return to classical mechanics, the

kinetic energy will disappear and the attractive interactions would be left

unopposed to squeeze the matter to non-existence.

In spite of the phenomenal quantitative successes of quantum mechanics in

passing all experimental tests (and there is a huge number of them), including

the daily flawless functioning of millions of devices based on it, still it is not

easy to ‘swallow’ its foundation, i.e., the particle/wave duality, the

probabilistic nature of its predictions, and the measurement effect (the collapse

of the wave function). Everyone would like to see a formulation compatible

with our perceptions which, nevertheless, would incorporate the observed

wave/particle duality. Such a formulation has not been achieved and not for the

lack of trying.

It is obvious that quantum mechanics is a vast field within Physics and a

presentation that does it full justice is way beyond the scope of the present

work. For my purposes, nevertheless, the above presentation of the basics of

quantum mechanics, albeit incomplete, suffices.

5.3.2 A Critique of Wendt’s View of Quantum Theory

I will now return to Wendt's views and present my critique of them, by means

of elaborating on a number of very specific shortfalls that Wendt's use of

quantum mechanics exhibits.

275

1. At the start of his brief exposition of quantum mechanics, Wendt makes the

following statement:

Quantum theory is perhaps best introduced by the classical worldview that it

overthrew. Like quantum metaphysics, the classical worldview is an interpretation

of physical theory, in this case classical physics, and as such essentially

metaphysical. It makes five basic assumptions: 1) that the elementary units of

reality are physical objects (materialism); 2) that larger objects can be reduced to

smaller ones (reductionism); 3) that objects behave in law-like ways (determinism);

4) that causation is mechanical and local (mechanism); and 5) that objects exist

independent of the subjects who observe them (objectivism?). In philosophy of

mind these assumptions are shared by materialists, dualists, and proponents of the

linguistic turn alike, and thus by extension by most positivists and interpretivists in

social science.

Quantum theory challenges all five. At the sub-atomic level physical objects

dissolve into ghost-like processes; wholes cannot be reduced to parts; the world

does not behave deterministically; causation is non-local; and objects do not exist

independent of the subjects who observe them. Importantly, these findings do not

necessarily invalidate the classical worldview at the macro level, since quantum

states normally ‘decohere’ into classical ones above the molecular level, which is

why the everyday world appears to us as classical. Decoherence has been a barrier

to developing a unified quantum theory encompassing both micro and macro

levels, and is a fundamental obstacle to the quantum consciousness hypothesis in

particular (see below). But at least at the micro-level the quantum revolution has

decisively overturned the claim of the classical worldview to provide a complete

description of reality. (Wendt 2006a: 190-191)

276

My criticism goes as follows:

(a) Atoms and electrons do not dissolve into ghost-like processes. On the

contrary, they become ‘more real’ as quantum science and technology

advance: one electron-transistor has been already obtained experimentally; the

electron-spin is manipulated so as to produce giant-magnetoresistance (Nobel

Prize 2007), a discovery that has already found practical use in modern

computers. One-electron spin is controlled by laser beams in order to act as

quantum bit (qubit) in quantum information processes; electrons in properly

designed metal-insulator interfaces have given rise to the new field of

plasmonics; coherent flows of electrons in superconductors exhibit interference

and quantization at a macroscopic scale; atoms are cooled down and produce

coherent beams exhibiting interference and a host of other impressive effects;

atoms have been photographed and manipulated one by one; etc. All these

phenomena are natural consequences of the fact that the atomic and the

subatomic wave-particles obey quantum laws. We can say that it is quantum

mechanics that allows us to test and confirm the reality of these microscopic

particles in so many and ever expanding ways.

(b) Wholes are reduced to parts. Reductionism is the day-by-day tool of

research physicists working within the framework of the atomic idea and

employing – of course – quantum mechanics.

(c) The question of determinism in quantum mechanics has two sides: the wave

function itself develops in time deterministically (the present determines the

future), until a measurement occurs. The measurement is essentially the

interaction of the quantum system (which obeys the fundamental uncertainty

principle) with an object obeying classical physics, i.e. one that is not subject

277

to the restrictions of the uncertainty principle (Landau & Lifshitz 1977: 2-4).

Thus, while measuring very accurately the position of a quantum particle, we

have to transfer to it a large amount of momentum so that the uncertainty in the

momentum becomes very large as dictated by the uncertainty principle. In

general, it can be said that a measurement acts as filter for the various

eventualities incorporated in the wave function: the ones which are inconsistent

with the outcome of the measurement are eliminated, while the ones which do

not contradict with the outcome are allowed to pass to the future. As a result of

this fundamental feature of the measurement in QM, the wave function

changes discontinuously as to be consistent with the result of the measurement.

The prediction of this outcome is in general of a probabilistic nature, for

example we can predict that the measurement of the energy of the carbon

atomic system being in a hybrid state will give this value with such a

probability. Of course, there are cases where the combination of wave function

and the measurement procedure is such that all the eventualities present in the

wave function are allowed to pass. In such cases we can predict the outcome

with certainty. This is why research physicists do not doubt the existence of the

electron.

(d) Classical physics is only a very satisfactory approximation to quantum

mechanics when certain limiting conditions are satisfied. The world is quantum

at all levels.

(e) Decoherence is a common wave phenomenon appearing within both the

classical formalism and the quantum reality; and it does not prevent quantum

mechanics to be a unified theory at all levels, micro-, macro-, mega-.

278

2. Furthermore, Wendt’s reference (2006a: 191) to the “four findings from

quantum theory: wave-particle duality, wave function collapse, the

measurement problem, and non-locality” needs, in my view, revisions because:

(a) Wave/particle duality is not a finding from quantum theory; it is an

undeniable experimental fact that had to be incorporated in any acceptable

physical theory and –of course- in the quantum theory.

(b) Wave function collapse and the measurement problem are essentially the

same thing; finally, the non-locality stems from the wave-like propagation of

wave-particles, i.e. from the wave /particle duality.

3. Wendt states also that

Wave-particle duality refers to the fact that sub-atomic phenomena have two

irreducible and non-equivalent descriptions. Under some experimental conditions

they are best described as waves, in others as particles. (Wendt 2006a: 191)

I would like to comment here that quantum mechanics is a complete self-

consistent theory, which does not need classical mechanics in order to produce

its quantitative predictions. In the cases where quantum mechanics is

satisfactorily approximated by classical mechanics, the use of classical

mechanics is preferred since the latter reduces significantly the calculational

effort.

4. Regarding Wendt’s statement that

279

Wave-particle duality challenges two assumptions of the classical worldview. One

is that science can achieve an integrated, unitary Truth about the world… The other

challenge is to the materialist view of matter. (ibid: 191)

I have the following remarks:

(a) Wave/particle duality and quantum mechanics do not challenge the first

assumption of the classical worldview. On the contrary, it goes a long way

towards making reality what was a dream before. This follows from the fact

that quantum mechanics extended the subject matter of physics from

macroscopic motions (to which classical mechanics is restricted) to all kinds of

microscopic motions and, as a result, to the structure and properties of protons,

neutrons, nuclei, atoms, molecules, solids, liquids, planets, stars, dead stars all

the way to the universe; it made also possible the emergence of cosmology as a

serious scientific subject.

(b) Quantum mechanics is based on the picture that the material world consists

exclusively of elementary point-like material particles which nevertheless

propagate as waves, while they retain their point-like property all along. By

accepting this picture (which does imply the Uncertainty Principle) we obtain a

tremendous amount of information and ‘understanding’ about the world. Here,

I find it appropriate to quote the brilliant physicist Richard P. Feynman:

If, in some cataclysm, all of scientific knowledge were to be destroyed, and only

one sentence passed on to the next generation of creatures, what statement would

contain the most information in the fewest words?

I believe it is the atomic hypothesis that all things are made of atoms–little

particles that move around in perpetual motion, attracting each other when they are

280

a little distance apart, but repelling upon being squeezed into one another. In that

one sentence, there is an enormous amount of information about the world, if just a

little imagination and thinking are applied. (Feynman 2007: 4)

5. Wendt continues the deployment of his ideas by presenting us with the

following:

Wave function collapse refers to the fact that the transition from wave to particle is

instantaneous in time and has no apparent physical cause. Such ‘quantum leaps’

challenge the determinism of the classical worldview, and as such have caused

much angst among physicists, with Einstein famously complaining that ‘God does

not play dice’. But their anomalous character also points toward a possible

solution, since wave function collapse is strongly analogous to our experience of

consciousness, which involves free will and also does not seem to have a physical

cause – an analogy that the quantum consciousness hypothesis will exploit.

The measurement problem refers to the fact that it is impossible to measure

quantum phenomena without disturbing them: the process of measurement

inevitably leads to a change in the appropriate description of sub-atomic particles.

As long as we don’t measure them they appear as waves, and as soon as we do as

particles. This challenges another basic assumption of the classical worldview, the

subject-object distinction, and with it the possibility, even in principle, of true

objectivity. In quantum measurement observer and observed initially constitute a

single system, rather than two as they are classically. Far from being just a given,

the subject-object distinction is now emergent from the process of measurement

itself, which makes a “cut” in a previously undivided whole. Within social science

post-modernists, feminists, and others have made similar critiques of the subject-

object distinction at the macro-level, but generally without a quantum basis. A

281

quantum connection would give these critiques additional force, and point toward

the necessity of a participatory epistemology in social inquiry. (Wendt 2006a: 192)

However, the wave function collapse, as a result of a measurement, does not

imply an instantaneous transition from wave to particle, as Wendt seems to

claim in the above passage. The particle per se remains unaffected by the

measurement; its state of motion as described by the wave function may

change abruptly from the wave function before to another wave function so as

to be consistent with the results of the measurement. The measurement of the

position does not mean that a wave is transformed to a particle. It simply

means that the change of the wave function to a small spot (as a result of the

position measurement) revealed the particle aspect which existed all along.

Wendt’s unconventional picture of the elementary particles seems to come

from his implicit identification of the measurement in general with the position

measurement; actually, in almost all cases, we measure other (than the

position) quantities (which are more relevant from the observation point of

view). (By the way, the connection of the wave/particle duality with some

feminists’ post-modern critiques of the subject-object distinction at the macro-

level might have remained unconceivable, had it not been for Wendt’s

idiosyncratic representation of quantum mechanics!).

6. Moreover, in the subject of Wendt’s arguments about quantum mechanics, I

cannot but point to the following statement since, in my view, it contains some

misconceptions:

282

Findings like these overthrew the classical worldview at the micro-level, but not at

the macro, where classical thinking still dominates. The reason is decoherence. As

we have seen measuring sub-atomic systems interferes with them, collapsing

quantum waves into classical particles. Importantly, this applies not just to

measurements in physicists’ laboratories, but in nature everywhere. Whenever

particles interact they are in effect ‘measuring’ each other, inducing decoherence.

That’s why in everyday life we see only material objects, not wave functions;

quantum effects quickly wash out beyond the molecular level, leaving classical

mechanics as the appropriate description at the macro. (ibid: 193)

First, there are macroscopic systems (consisting of interacting microscopic

particles) which are coherent and –as a result- exhibit interference and

quantization phenomena. Liquid Helium 4 below the temperature of 2.18º K is

one example; electric currents in superconductors are another - there are

commercially available devices called SQUIDs (Superconducting Quantum

Interference Devices) which are based on macroscopic coherent systems. The

reasons that ordinary macroscopic systems decohere is not due to the

interactions among their constituent particles. Interaction with the environment

is the main cause of decoherence. There are also other cases where the direct

test of coherence cannot take place. E.g., in the case of ordinary solids, the

possibility of the interference test is a priori excluded, since they cannot both

occupy the same volume at the same time.

7. It is self-evidently incorrect to identify the process of measurement in

general with the emergence of decoherence, as Wendt does in the following

sentences:

283

Quantum brain theorists are trying to bridge the yawning gap between sub-atomic

particles and the whole brain in such a way that quantum coherence might be

transferred from the former to the latter. The key problem is identifying physical

structures in the brain whose properties will insulate particles from measuring (and

thus collapsing) each other, while simultaneously enabling them to be entangled

(and thus having coherence). (ibid: 193)

8. Furthermore, Wendt states that:

For this purpose they are using quantum field theory – a generalization of quantum

theory that deals with macroscopic phenomena – to model brain activity. (ibid:

194)

It must be clarified with respect to the above that quantum field theory

completes quantum mechanics by adding particle properties to classical waves.

Thus, quantum field theory is a unifying quantum formalism which treats on

equal footing both ‘material’ (e.g. electrons) and ‘interaction’ (e.g. photons)

particles. It is a convenient formalism for many-body systems.

9. Moreover, Wendt’s attempt to connect quantum mechanics with panpsychist

ontology lacks clarity. The leap towards which Wendt aims cannot be

adequately substantiated by an under-elaborated statement of intent:

For that we need to replace the materialist view of matter that underlies classical

approaches to the mind-body problem with a panpsychist ontology, which is the

284

view that something like human consciousness goes all the way down to the sub-

atomic level. (ibid: 195)

10. Furthermore, in the same page Wendt goes on to argue that:

The question of how consciousness emerges from matter is therefore spurious,

since in some sense it is there all along. What is emergent is rather the distinction

between consciousness and purely physical matter from an underlying reality that

is neither. David Bohm (1980) calls this underlying reality the ‘implicate order,’ as

distinct from the ‘explicate’ order of physical matter and consciousness. As such,

panpsychism might also be described as a ‘dual aspect’ or ‘neutral’ monism. (ibid:

195)

The above passage, besides being subject to the claim that it lacks clarity as

much as the excerpt in point 9 above, also leads to questions such as “Has an

electron consciousness? In what sense? Has quantum mechanics anything to do

with these questions?”. These questions are never addressed or elaborated,

even for the sake of clarifying whether they are worthy of elaboration.

11. Wendt goes on to connect mental processes such as consciousness, or the

lack of it, with coherence and decoherence, respectively. This, taken at face

value, ought to have one conclude that a Helium 4 nucleus (the so-called alpha

particle) has consciousness, since it is in a coherent state:

285

But are we then to believe that rocks have consciousness? No, because of

decoherence. From a quantum perspective, part of what constitutes life may be the

ability to maintain coherence in a multi-particle system. (ibid: 196)

Coherence, or the lack of it, is not the basis for distinguishing living systems

from the non-living ones. Both, since they are multi-particle systems held

together mainly by electromagnetic forces, are generally in an incoherent state

as a result of their energy-level spacing being much smaller than the interaction

energy with the environment. This does not exclude the possibility that some

particular subsystems of a biologically advanced species may be in a coherent

state, at least for a short time. Because they lack experimental or detailed

theoretical support, such speculations, which have been proposed in the

literature, are not generally accepted by the relevant scientific community.

12. Wendt’s 2010 paper represents his version of a ‘quantum social science’

along the lines that had already been laid out in his earlier work (Wendt

2006a). Were it not for the ‘holographic principle’, which Wendt invokes here

in order to supply his analysis of the international political system with a new

theoretical tool, we shouldn’t have added anything more to the previous critical

comments on Wendt’s misunderstanding of quantum mechanics. Although I

find a direct ‘transplantation’ of the ‘holographic principle’ from Physics into

the realm of IR theorising incomprehensible, Wendt argues to the contrary; that

is, he argues for its adaptation in IR Theory as a valid scientific – and not

merely a metaphorical – analogy:

286

The holographic principle does not depend on any particular medium for its

realization and as such is highly general. It has been argued by some cosmologists,

for example, that the entire universe is a holographic projection, not just

metaphorically but literally (for example, Susskind 1995). This puts us in a good

position to extend the holographic perspective to IR, where the mechanisms and

medium go beyond mere visual perception of physical objects. (Wendt 2010:

290)180

One might respond to this claim by taking refuge in Susskind’s own words

when explaining to the Scientific American’s audience what this principle is all

about:

Question: If I get you correctly, the holographic principle extends the

complementary model of a black hole to the universe.

180 In fact, in order to clarify things, I should say that Wendt’s (2010) enrichment of his earlier ‘quantum consciousness hypothesis’ with the inclusion of the ‘holographic principle’, was due to the inspiration he drew from the work of the neurophysiologist Karl Pribram on ‘visual perception and the holographic brain’:

The conventional view of visual perception is representational: light refracted of external objects pours via the eyes into the brain, which uses the information in the light to form cognitive representations of the objects. This view is very classical in its assumption that there is a clear ontological distinction between subject and object, and that perception takes place locally, within the brain. In contrast, the quantum consciousness hypothesis suggests a holographic model of perception, which was first put forward in the 1960s by Karl Pribram, who was trying to explain why we experience objects as ‘out there’ in the world even though their images are actually on our retinas. (Wendt 2010: 287, the italics are mine).

The idea that Wendt found so attractive in Pribram’s works (Pribram 1971, 1986) is that in the brain, just as within a holographic image, each part contains the whole within it; an interesting, however controversial, scientific image of the functioning of the human brain. The idea of the ‘holographic universe’, as that has been framed by the distinct works of Pribram and the renowned physicist David Bohm, is also drawn upon by the widely discussed book of Michael Talbot, The Holographic Universe: The Revolutionary Theory of Reality (Talbot 2011). However, this work, which explores the metaphysical implications of quantum physics and argues that a holographic model of the universe could explain supersymmetry, various paranormal and anomalous phenomena and can act as the basis for mystical experiences, seems to belong, as many others of the same sort, to the so-called ‘quantum mysticism’ literature. As such, it probably has no place in any serious discussion about either modern interdisciplinary brain/consciousness studies or the metaphysical implications of quantum theory.

287

Answer: Yes. Suppose we want to describe some system with enormous precision.

To probe with great precision, you need high energy. What’s eventually going to

happen as you try to get more and more precise is you’re going to start creating

black holes. The information in a black hole is all on the surface of the black hole.

So the more and more refined description you make of a system, you will wind up

placing the information at a boundary.

There are two descriptions of reality: either reality is the bulk of spacetime

surrounded by the boundary, or reality is the area of the boundary. So which

description is real? There is no way to answer it. We can either think of an object

as an object in the bulk space or think of it as complicated, scrambled collection of

information on the boundary that surrounds it. Not both. It’s an incredibly

scrambled mapping of one thing to the other thing. (Susskind 2011: 67)181

In concluding this section, I would like to stress that quantum mechanics is a

physical theory and, as such, it must be judged by its ability to account for as

many observations and experiments as possible in all scales of the natural

world and the man-made structures. Quantum mechanics is phenomenally

successful in this respect, having extended the subject matter of physics from

the classical realm of studying macroscopic motions to almost everything from

181 As Susskind argues, his ‘holographic principle’ is an extension of the complementary model of a black hole to the universe. The more deeply one conceives the latter, the better he understands the former. The American physicist explains the complementary model of a black hole as follows (Susskind 2011: 66-67): “It turns out that the mathematics of the event horizon of a black hole is very similar to the uncertainty principle. Again, it’s a question of ‘or’ versus ‘and’. At a completely classical level something falls into a black hole, something doesn’t fall into a black hole, whatever. There are things outside the black hole, and are things inside the black hole. What we learned is that’s the wrong way to think. Don’t try to think of things happening outside the horizon and things happening inside the horizon. They’re redundant descriptions of the same thing. You describe it one way, or you describe it the other way. This means we have to give up the old idea that a bit of information is in a definite place [see “Black Holes and the Information Paradox,” by Leonard Susskind; SCIENTIFIC AMERICAN; April 1997]”. One without advanced technical knowledge of Mathematics and Physics may also consult Chapter 18 (“The Universe as a Hologram”) of Susskind’s 2004 book, The Black Hole War – My battle with Stephen Hawking to make the world safe. On the other hand, a mathematical physicist may enjoy reading Roger Penrose’s brilliant exposition of the ‘holographic principle’ in his masterpiece, The Road to Reality (Penrose 2004: 920-923).

288

the quarks to the Universe. However, the methodology of Physics (controlled

experiments, quantitative character, use of mathematical language to derive

results starting from a few basic relations) imposes severe restrictions

regarding which systems can be studied profitably by physical methods. For

instance, physicists have just started to calculate large biological molecules

which participate in the structure and the function of cells. Convincing

calculations of more complex systems such as a cell are out of reach for the

time being. This is obviously more so for the human brain and, certainly, for

the structures and functions of the human societies. Any approaches to these

subjects using quantum mechanics, including Wendt’s paper, cannot be

considered scientific statements/conjectures in the Popperian sense, simply

because they are not falsifiable.182

5.4 The Mind-Body Problem in Wendt’s Theory

The motivation behind Wendt's embarking on his quantum programme towards

a quantum social science originated in the mind-body problem. It is common

knowledge that there have been various philosophical angles on this problem

since Descartes, whose writings shaped the way the problem is understood in

modern and contemporary philosophy. One comes across monistic approaches,

whether idealist or materialist (physicalist), dualistic approaches, interactionist

or not, functionalist approaches, etc.183 Wendt does not express any, even

programmatic, preference among all these. Instead, he claims that all these

approaches ultimately suffer from one common shortcoming: they are based on

the world-view suggested by classical physics. Then he goes on to indicate 182 See Popper ([1957] 1997), as well as Chapter 1 (“Science: Conjectures and Refutations”) and Chapter 10 (“Truth, rationality, and the growth of scientific knowledge”) in Popper ([1963] 1996) .183 See, for example, Kim (1998).

289

how, in his view, a quantum approach to consciousness can overcome this

hurdle and solve the mind-body problem.

But this is misleading on several scores. To begin with, there does not seem

to be a unique world-view bequeathed to us by classical physic. 184

Furthermore, the various quantum approaches to consciousness entertain

different philosophical background assumptions as to how the mental is

‘related’ to the material or the physical. I will now take up these points in turn.

As any serious student of the history of physics will attest, the development

of thermodynamics and electromagnetic theory during the 19th century guided

physics toward a rejection of the mechanistic world-view that dominated the

17th century Scientific Revolution. In fact, it is normal to maintain that classical

physics has bequeathed to us two radically different metaphysical schemes.

The atomistic metaphysics recognizes a plurality of fundamental substances

with diachronic identity (continuants), which is seated in their ‘primitive

thisness’185 or ascertained by appeal to the continuity of their spatiotemporal

trajectories. By contrast, the plenum metaphysics recognizes a single

fundamental substance (‘ether’), which can manifest quality complexes with no

diachronic identity (ephemerals). Discreteness and continuity are, respectively,

the cardinal images. Particle theories and field theories, with Newtonian

184 Or, even if there were just one, Wendt’s depiction of it would not be entirely accurate. For instance, Wendt (2006: 190) claims that one of the basic tenets supporting what he takes to be “the classical world-view” is that “objects behave in law-like ways (determinism)”. Of course, a world bound by laws of nature need not necessarily be a deterministic world. In addition, it has been established that even the world of Newtonian particle mechanics and gravitation is not a safe haven for determinism. On such matters see Earman (1986). 185 Primitive thisness is meant as a principle of individuation that transcends an entity’s properties. The traditional term, derived from Scholastic vocabulary, is ‘haecceity’.

290

mechanics and Maxwellian electromagnetism as the corresponding exemplars,

have served to typify the atomistic and the plenum metaphysics respectively.186

None of these metaphysical schemes has survived, intact or unaltered, the

advent of quantum theories. The notorious ‘wave-particle duality’, manifested

by quantum systems, is actually an attestation of this fact. Of course, this

diagnosis of the problem differs radically from the one offered by Wendt:

Wave-particle duality challenges two assumptions of the classical worldview. One

is that science can achieve an integrated, unitary Truth about the world. Quantum

theory seems to be true, but its truth requires contradictory narratives … The other

challenge is to the materialist view of matter. (Wendt 2006a: 191)

With respect to what Wendt cites as the first challenge, the expression ‘wave-

particle duality’ refers to the capacity of a single quantum mechanical system

to manifest behaviour similar to that of classical waves or to that of classical

particles under diverse experimental conditions. The possession of this capacity

prohibits the possibility of giving a unified description of quantum entities with

classical concepts or images. Still, there might be such a unified description

that relies on non-classical concepts or images. Much of the philosophical

work on quantum theories aims at articulating precisely such a description.187

What’s more, the thesis that science can achieve an “integrated, unitary Truth

about the world” is not, properly speaking, part of any physical theory, whether

186 This sketch of the distinction is oversimplified on several scores. For one thing, ether theories have not invariably been embedded into a monistic ontology; dualistic theories (ether plus material particles) abound in the repertoire of classical physics.187 Admittedly, this task has not yet been achieved. Moreover, many interpreters of quantum theories do not share the commitment to such a task. Notably, Bohr ([1949] 1983) insisted that the concepts of classical physics are indispensable for the description of experiments on quantum systems.

291

classical or not. It is, rather, part of a realist stance toward scientific theories.

And almost every physical theory, including quantum mechanics, is subject to

both realist and antirealist readings.188

Turning now to the variety of philosophical background assumptions

embraced by the various quantum approaches to consciousness, one can readily

discern the following distinction (cf. Atmanspacher 2006). Some of those

approaches aim at explaining the – at least apparent – interaction between

mental and material entities by positing some direct relation between the two.

In general, this relation might be one of reduction, emergence, or

supervenience supporting either a monistic or a dualistic ontology. Other

approaches posit indirect relations between mind and matter mediated by some

third, psychophysically neutral, category of entities. On such approaches, mind

and matter are conceived as dual aspects or effects of a common underlying

reality.

Of course, both the direct-relation and the dual-aspect scheme draw their

ancestry from the history of Western philosophy long before the advent of

quantum physics. Descartes and Spinoza, respectively, defended such schemes

– to mention just two of the ‘greats’. At this point, it is useful to examine how

various quantum approaches to consciousness stand vis-à-vis this distinction.

On the one hand, Wigner ([1961] 1983) appears to suggest a direct-relation

dualistic scheme. On the other hand, Stapp (1999: 159) talks about a ‘hybrid

ontology’ which possesses ‘both idea-like and matter-like qualities’ and, as a

consequence, intimidates a dual-aspect scheme. By contrast, the approach

elaborated by Vitiello (1995, 2001, 2002) belongs to the direct-relation 188 Again, the relevant literature contains numerous arguments to the conclusion that quantum mechanics signals the demise of realism. But these arguments rest on narrow construals of what realism amounts to.

292

physicalist variety. And the same is true for that proposed by Penrose (1989,

1994a, 1997a and 2011). Lastly, Bohm ([1980] 1997: 209) affirms explicitly

the dual-aspect orientation of his approach by writing: “… we do not say that

mind and matter causally affect each other, but rather that the movements of

both are the outcome of related projections of a higher-dimensional ground”.

In conclusion, Wendt’s programme towards a quantum social science lacks

specific sound metaphysical underpinnings. Its metaphysical assumptions

concerning matter conceal a number of conceptual confusions; moreover, its

metaphysical assumptions concerning the relation of mind to matter remain

promiscuously eclectic.

5.5 The Obstacles Towards a New Science of International Relations

As should be clear by now from the aforementioned analysis, Wendt (2006a)

and Wendt (2010) are the first, and so far the only, steps of his new project in

development, Quantum Mind and Social Science. In that, they represent the

direction he intends to follow in the process of actually writing it up.

Wendt seems to think now that his previous two efforts to combine

positivist and interpretivist methods of inquiry in social science and IR theory

through the means of his ‘via media’ (Wendt 2000) proved to be less

successful than initially hoped. This is mainly due to the remaining unsolved

problem of consciousness which is directly connected to that of the mind-body

(in the Cartesian dualistic scheme) and comprises the essence of human

behaviour as the basis for any reasonable social science. Wendt undertakes the

enterprise of building a new quantum social science, not in order to pass this

293

issue by or transcend it, but in order to resolve it via his ‘quantum

consciousness hypothesis’ (QCH).

As discussed in the previous chapter, Wendt’s first effort had the ‘agent-

structure issue’ as the point of its departure (Wendt 1987). In order to resolve

this ‘open problem’, Wendt brought SR to the fore, because of the following

three assumptions which SR entailed and which he considered to be connected

with the purposes of his overall venture: 1) there is a real natural and social

world out there, independent from us and our ways of knowing it, 2) scientific

theories have to be taken at their face value until they become falsified and 3)

reality consists of both observable and unobservable entities. Due to the third

assumption, Wendt felt it was legitimate to bring ideas into the existing and

inaugurate discussions on causes and reasons in IR theory, by substituting the

respective notions of explaining and understanding by the new term of ‘mutual

constitution’, namely making feasible a synthesis of the two of them. In this

vein, SR gives to ontology a priority over epistemological and methodological

concerns, putting the latter aside and functioning as the single meta-theoretical

framework able to underpin any IR theory which aims to bridge the opposing

epistemological traditions of positivism (the current mean-stream epistemology

in the USA) and interpretivism (mainly, however not exclusively, developed in

continental Europe). In other words, the so-called ‘first Wendt’ “focused on

structurationism and scientific realism as ways of thinking about IR” (Keeley

2007: 417).

On the other hand, the ‘second Wendt’, as deployed in his STIP and other

relevant publications, seems to put an emphasis on the state and the

international system, although Wendt himself denies it (Wendt 2000: 174-175),

294

when arguing for his having presented states in STIP only as quasi-

autonomous, self-organising actors. My impression, however, is that the state

in STIP “is generally treated as a unitary and intentional actor” (Keeley 2007:

420), something which brings Wendt closer to realist and liberal IR scholars

who conduct their empirical research within a positivist epistemology and

which simultaneously fills radical interpretivists of various sorts with

frustration, since they are unambiguously inclined towards denying any

possible reification of the state and the international system. This shift towards

an enhanced importance of the state in Wendt’s theorizing became apparent

with the appearance of his paper, “The state as a person” (Wendt 2004). It is

here that Wendt takes on board the notions of supervenience and emergence in

order to approach the state as an organism (more accurately as a

superorganism) with collective consciousness. Wendt’s most recent ‘quantum

consciousness hypothesis’ is traceable to the aforementioned idea of the

collective consciousness (of the state), which he presumably felt the need to

further explore and elaborate. All in all, the ‘second Wendt’ is an outspoken

supporter of a ‘middle-ground’ constructivism which is based “on an uneasy

tension between mental causality and rump materialism that shows itself as a

conflict between upward determination of ideas and their downward causation

on the material world” (Kessler 2007: 243).

At this point, it is worth referring to Kessler’s well presented argument

concerning the ‘inefficiency’ of the ‘supervenience hypothesis’ to

explain/solve the mind-body problem, if only for the sake of its simplicity and

clarity:

295

According to supervenience, the argument goes, mental event A is supervenient on

physical event C (Ego), and  is supervenient on Ĉ (Alter).

What we are interested in is how A can influence the physical basis of Ĉ as to

allow for  to emerge. But the step from A to  can only be explained in two

ways: either we argue that the change from A first requires a change on this

physical basis C. Ego’s C now causes Alter to change his brain into Ĉ which then

leads to Â. Or we argue that A caused  directly. The first option is unable to

explain mental causality, since the real ‘cause’ is from C→Ĉ, a form of reductive

materialism. The second option, that is an explanation in the form of A→Â falls

back to ontological dualism.

This inability arises from a genuine conflict between upward determination and

downward causation which has not yet been resolved. This is what I regard to be

the central problem: by taking mental causality seriously, supervenience arguments

are bound to collapse into either ontological dualism, raising again the issue of

bridge-laws, or materialism proper. Now, one could argue that this problem might

not be a problem at all, since they can both be considered valid for they simply

highlight different aspects and simply see different things. This is the dual

explanation strategy (Wendt 1995). This avenue is a recurrent theme in Wendt.

(Kessler 2007: 256)

As a consequence, Wendt (2006a) develops his QCH as a possible solution to

the mind-body problem, via the inspiration he draws from the theoretical work

on the ‘consciousness issue’ which has been produced by Penrose (1994),

Hameroff and Penrose (1996) and Hameroff (2001).

There are two main issues with this approach. The first has to do with

Wendt's understanding of Penrose's work on the ‘consciousness issue’ and the

insufficient coverage that Wendt gives to the specifics of a theory, which is

296

important for his project. The second has to do with the incorporation of his

view that human consciousness is seen as a quantum mechanical phenomenon

in that it behaves as both a wave and a particle, in compliance with the wave-

particle duality of quantum mechanics, as a gateway to the solution of the

problems he tries to tackle in QCH. In the former case, an inaccurate

understanding of the specific theoretical work of Penrose's undermines Wendt's

project at its roots. In the latter case, a misplaced analogy between the way

human consciousness works and Quantum Theory in his QCH undermines the

structure of Wendt's project.

5.5.1 Penrose's Theory on the ‘Consciousness Issue’

It is remarkable that Wendt, in both his first piece and his most recent pieces

(Wendt, 2006a, 2010), is very eclectic with regards to the use of the very wide

bibliography produced by Hameroff, Penrose and their critics on the quantum

consciousness issue, confining his references only to the following three

pieces: Penrose (1994), Hameroff and Penrose (1996) and Hameroff (2001).

More indicative is, of course, the absence of any critical approach to Penrose’s

ambitious plan to ground a theory of quantum consciousness, launched by his

book, The Emperor’s New Mind – Concerning Computers, Minds and the Laws

of Physics (Penrose 1989).

The book is an attack on Artificial Intelligence (AI), since its purpose is to

demonstrate that the human mind cannot be modeled by a Turing machine or

any other mechanical devise. In trying to explore the nature of mathematical

creativity, he guides us in the labyrinths of Mathematics (here one finds a good

introduction to lambda calculus and to the Mandelbrot set, among others),

297

Mathematical Logic (with an interesting but controversial discussion of Gödel’s

Theorem, which was bound to cause great debate about its ‘consistency’ after

the publication of the book) and modern Physics (ranging from quantum

mechanics and quantum gravity to the discussion of Time’s arrow). It is only

within the last chapter of the said book that he comes to grips with the idea of

quantum consciousness, since he is basically captured by the quest of “where

the physics of the mind lies”.

This book sparked the appearance of long, detailed and profound reviews on

its subject matter, most of which have been positive, in general terms, vis-a-vis

his exposition of Mathematics and Physics (Dewdney 1989, Barr 1990,

Senechal 1992), but have ignored or been very skeptical of his ‘quantum

consciousness hypothesis’ (Johnson 1990). It is important to cite Johnson at this

point, since he first explains, very clearly, how Penrose connects quantum

physics with the problem of consciousness and then goes on to aptly criticize it:

Rather, he [Penrose] seems to believe that consciousness somehow arises from this

interaction between the quantum realm and the macroscopic world of everyday

existence, where the traditional rules of billiard-ball mechanics hold sway. When

we are thinking about a problem, he proposes, the many possible solutions exist

simultaneously in quantum superposition, much like the myriad positions of an

electron around a nucleus. By considering all the possibilities at once, the brain can

deal with incredibly complex problems beyond the power of algorithms. At some

point in these quantum cogitations, the probability wave collapses in a sudden

flash. We see the truth as the answer to the problem crystallizes like ice forming on

a pond. We have an epiphany – contact with Plato’s world.

298

But here, just as Penrose’s story is getting good, it comes to a screeching halt.

Try as he might, he can find nothing in neuroscience to suggest that quantum

behavior has anything to do with the way the brain works. Small as it is, a single

neuron is enormous compared with an atom; its behavior seems best described by

the macroscopic laws of chemistry. And even if there were some kind of quantum

computer in our heads, what would cause all of its simultaneous calculations (or

whatever they are) to collapse suddenly into an insight?

Penrose says the answers to such questions will have to await the completion of

his current effort to reconcile quantum theory with general relativity, which

explains things such as stars. For most of this century, physicists have tried in vain

to combine these theories of the very small and the very large. This joining of the

microcosm and the macrocosm into a theory of everything would explain the

quantum-mechanical paradoxes that seem to occur when we intrude on the

subatomic realm. If Penrose is right, it would also explain the universe inside our

heads, revealing how brains intersect with the Platonic world. (Johnson 1990: 48-

49)

Penrose tried to reply in detail to his critics by writing a new book, Shadows of

the Mind – A search for the missing science of consciousness (Penrose 1994a),

a sequel to The Emperor’s New Mind, in which he further developed his

argumentation on three levels: First, in order to take into consideration the

counter-arguments that appeared after his first book has been published, he

rewrote his Gödel arguments and came up with a new one, which, according to

him (Penrose 1994b), makes a much stronger case against any computational

picture of consciousness. His new Gödelian argument seems to be in agreement

with the famous and controversial use by John Lucas of Gödel-type arguments

against the computational model of mind (Lucas 1961, 1970), as is clearly

299

stated in another, more recent, paper (Penrose 1997). Moreover, if one finds the

presentation of this argument in Shadows of the Mind fairly long and involved,

I may call attention to another presentation of it by Penrose himself (Penrose

1995), where he set out the argument only in its essentials. Searle has also

provided us with his own version of, as he calls it, “Penrose’s version of

Turing’s version of Gödel’s incompleteness proof”, which Penrose uses to try

to prove Lucas’s claim (Searle 1995: 96). Second, as concerns the discussion of

Physics in Shadows of the Mind, the author’s view of the quantum state

reduction has somewhat shifted from that which he expressed in The

Emperor’s New Mind. Third, after Hameroff wrote to him about his work on

microtubules and their role to control the strength of synapses, Penrose felt

more comfortable with this news than with his initial thought cast in his first

book and based on the idea that nerve signals could really be treated quantum

mechanically. In his new treatise he seems to be more confident that quantum

effects could actually be happening in the brain and this is pictured in the title

and context of Chapter 7 of Shadows of the Mind which has the telling title

“Quantum Theory and the Brain” (Penrose 1994a). It is here that Penrose

induces us to study the role of cytoskeletons and microtubules in correlation

with that of quantum coherence as a new means of building up a quantum

model of the brain and thus explaining consciousness (relevant are also

Penrose 1994c, Hameroff 1994, Hameroff and Penrose 1996, Hameroff 2001).

This new book by Penrose has sparkled a great debate about many different

aspects of its context; in fact, mathematicians and logicians (e.g., Feferman

1995, McCullough 1995, Faris 1996, Odifreddi 1997, Franzén 2005),

physicists (e.g., Josephson 1994), cognitive scientists (e.g., Klein 1995, Baars

300

1994, 1995, Bringsjord & Xiao 2000), computer and robot scientists (e.g.,

Moravec 1995, McCarthy 1995, McDermott 1995, Aaronson 2013, chapter

11), philosophers (e.g., Searle 1995, Chalmers 1995, Maudlin 1995, Copeland

1998, Detlefsen 1998, Shapiro 1998, Berto 2009, Tieszen 1996, 2011) and

freelance science writers (e.g., Brown 1994), alike, have all participated in this

vivid dialogue through publications in journals, in order to discuss Penrose’s

version of Gödel’s incompleteness theorem (namely its ‘soundness’ as to its

implications that, first, human reasoning cannot be formalized and, second,

there is a limitation on what computers can accomplish relative to humans), as

well as his interpretation of quantum mechanics and, subsequently, its

relevance (or not) to the ‘laws of brain’ and consciousness. Penrose has replied

to the scientists who have submitted commentaries on Shadows of the Mind in

the framework of a forum organized on the said book by the electronic journal

PRSYCHE (Penrose 1995).

The first issue at stake of this debate concerns the implications of Penrose’s

version of Gödel’s incompleteness theorem for the capacities of human

reasoning vis-à-vis those of computers. With regards to this, one could present

an argument in full alignment with Detlefsen according to which, like Lucas’

and others’ before it, Penrose’s argument “fails to establish any inherent

asymmetry between the capacities of humans and machines as regards their

abilities to evaluate their own beliefs” (Detlefsen 1998: 135). This happens

because

When we say, then, that we can judge our own beliefs to be consistent, we do not

mean this in such a way as to imply that the set referred to as ‘our beliefs’ should

301

be taken to include this judgment itself. We do not clearly have a capacity to judge

sets of our beliefs containing their own consistency judgment to be consistent. The

fact that […] various computing machines lack this capacity does not therefore

point to a difference between them and us. (Detlefsen 1998: 136)

Many of the aforementioned papers on Penrose’s and Lucas’ arguments about

Gödel’s incompleteness theorem are so technical that it is difficult for the

general audience to follow the full force of the arguments presented in them.

Given the above, it is strikingly interesting to note that many years ago the

philosopher J. L. Mackie commented on Lucas’ claim in his book, Ethics:

Inventing Right and Wrong:

An ingenious argument against determinism, based on a theorem about the

necessary incompleteness of a certain kind of mathematical system, has been

developed and defended especially by J. R. Lucas. I shall not attempt either to

expound or to criticize it here, but shall bluntly state my opinion that what it shows

is not that human minds are not deterministic structures, but at most that some

human minds are not closed deterministic structures, that certain sorts of

provocative input can change the way they work; but we knew that already.

(Mackie 1990: 219)

However, I think the most clear-cut and profound philosophical critique of

Penrose’s new argument about Gödel’s incompleteness theorem has been

offered by Tieszen (1996, 2011). This critique is demonstrated in a coherent

and most comprehensive way in the following passage, where Tieszen first

explores Gödel’s and Penrose’s differing beliefs regarding the nature of mental

302

and brain processes, and then steadily builds his own argumentation upon this

assumption. Here the fallacy of the interconnection of Gödel’s incompleteness

theorem with Penrose’s conjecture about a non-computational (possibly

quantum) nature of the human consciousness is evident:

Gödel once suggested that the argument that mental processes are mechanical is

valid if one assumes that (i) there is no mind separate from matter and (ii) the brain

functions basically like a digital computer (see Wang 1974: 326). Gödel evidently

thought that (ii) was a prejudice of our time that might actually de disproved. It is

interesting to compare this with the position of Roger Penrose in his book Shadows

of Mind (Penrose 1994a; see also Feferman 1996). Unlike Gödel, Penrose denies

(ii). He holds that the missing science of consciousness is to be a form of

neuroscience that recognizes non-computational brain processes. Penrose thinks

that Gödel’s incompleteness theorems imply some kind of Platonism. He also uses

the incompleteness theorems to argue that there are non-computational procedures

for knowing mathematical truths, but this is immediately equated with the view that

there must be non-computational brain processes (Penrose 1994a). It is fair to ask

Penrose the question that immediately gets us into a corner: how could the brain be

stimulated by abstract objects or, rephrasing it somewhat, what could the

relationship between the brain (which for Penrose is not a computer) and abstract

objects be? Given what a brain is and given the kinds of physical interactions in

which it can be involved, and given that an abstract object is typically understood

to be a non-physical object of some kind, it would be a category mistake to suppose

that brains could have access to abstract objects. One might expect to address

issues about non-computational mental processes, where these are not immediately

equated with brain processes, but there is no phenomenology of consciousness of

abstract objects. There is neuroscience with its “natural attitude” on one side, and

303

Platonism on the other, and not the slightest hint how the two could be related. One

might therefore think that the missing science of consciousness cannot consist only

of neuroscience (see Tieszen 1996 and 2005). (Tieszen 2011: 131)

This is not the place to present and develop the rich and substantial exchange

between specialists on such grand and, most importantly, contested scientific

and philosophical issues as the above. Suffice it only to cite the second part of

the closing paragraph of Faris’ piece, since it refers to Penrose’s speculations

about an allegedly “quantum nature and functioning” of human consciousness:

Finally, there is the proposal of a link between quantum mechanics and

consciousness. Penrose goes out of his way to admit that the evidence for this is

practically nonexistent. As he says, his argument requires that “our brains have

somehow contrived to harness the details of a physics that is yet unknown to

human physicists”. So we must take it as vision of a possible future for science. It

is hard to see the appeal of this vision. We do not understand either quantum

mechanics or the mind, but this does not suggest that one is the solution to the

other; most likely each will bring its own surprise. (Faris 1996: 208)

In order to put all the above in a nutshell and give a more acute and

compelling criticism of Penrose’s ‘quantum consciousness hypothesis,’ Ι quote

Michael Shermer’s relevant words in his regular column, “Sceptic”, in

Scientific American, where he cleverly invokes a good comment made by the

physicist Victor Stenger on the issue at stake:

304

The attempt to link the weirdness of the quantum world to mysteries of the macro

world (such as consciousness) is not new. The best candidate to connect the two

comes from University of Oxford physicist Roger Penrose and physician Stuart

Hameroff of the Arizona Health Sciences Center, whose theory of quantum

consciousness has generated much heat but little light. Inside our neurons are tiny

hollow microtubules that act like structural scaffolding. Their conjecture (and that's

all it is) is that something inside the microtubules may initiate a wave-function

collapse that results in the quantum coherence of atoms. The quantum coherence

causes neurotransmitters to be released into the synapses between neurons, thus

triggering them to fire in a uniform pattern that creates thought and consciousness.

Because a wave-function collapse can come about only when an atom is

"observed" (that is, affected in any way by something else), the late neuroscientist

Sir John Eccles, another proponent of the idea, even suggested that "mind" may be

the observer in a recursive loop from atoms to molecules to neurons to thought to

consciousness to mind to atoms....

In reality, the gap between subatomic quantum effects and large-scale macro

systems is too large to bridge. In his book The Unconscious Quantum (Prometheus

Books, 1995), University of Colorado physicist Victor Stenger demonstrates that

for a system to be described quantum-mechanically, its typical mass (m), speed (v)

and distance (d) must be on the order of Planck's constant (ħ). "If mvd is much

greater than (ħ), then the system probably can be treated classically." Stenger

computes that the mass of neural transmitter molecules and their speed across the

distance of the synapse are about two orders of magnitude too large for quantum

effects to be influential. There is no micro-macro connection. (Shermer 2004: 38)

One has also to underline at this point that Penrose seems to have neither

changed his own views about these issues nor come up with new ideas or any

305

evidence in support of his ‘quantum consciousness hypothesis’ in his most

recent writings which appeared after the publication of Shadows of the Mind

(Penrose et al. 1997, 2011). However, an important feature of The Large, the

Small and the Human Mind (Penrose et al. 1997) is the inclusion of short

critical essays by Abner Shimony, Nancy Cartwright and Stephen Hawking, as

well as Penrose's response to his critics in the final chapter of the book. It is my

belief that the most acute observations on Penrose’s attempt to connect Gödel’s

Theorem to the problem of human understanding are put forward in Shimony's

essay. Shimony’s main argument is nicely synopsized by Michael Redhead as

follows:

In his comments Shimony accuses Penrose of trying to ‘climb the wrong

mountain’, with his Gödelian argument. Shimony believes the distinctive features

of mentality are much more decisively brought out by arguments such as John

Searle’s ‘Chinese room’ argument, which shows, says Shimony, that correct

computation by an automaton does not constitute understanding, without any need

to appeal to complexities of mathematical logic. Shimony goes on to sketch a

metaphysics that he calls ‘modernized Whiteheadianism’, in which all physical

processes involve an element of protomentality. Penrose, in his reply, admits that

some such framework probably expresses best his general point of view. (Redhead

2000: 917)

To summarize, the fact that Wendt builds mainly (although not exclusively) his

own new theoretical approach to social science and international politics upon

Penrose’s work on quantum consciousness (Wendt 2006b), necessitated an

elaboration of Penrose's ideas along with some counter-arguments put forward

306

by Penrose's main critics. Wendt has, so far at least, taken Penrose’s theory, or

what he understands as Penrose's theory, for granted and treated it as

conclusive. A reason for this may be that Wendt uses popular science books on

quantum mechanics in his attempt to understand this physical theory.

However, it is a truism among physicists, mathematicians and philosophers

of science, alike, that without both a profound comprehension of the conceptual

tools and a technical knowledge of the mathematical ‘skeleton’ of quantum

mechanics, any attempt to interpret this theory is bound to be faulty with

regards to issues of methodology and metaphysics, as well as with regards to its

ability to describe precisely the physical world.

5.5.2 Quantum Consciousness and the Holographic Model in Wendt’s new

IR Theory

Having clarified the superficial treatment that Wendt makes of Penrose's

theory, it is now time to turn my attention on the use that Wendt makes of the

view according to which, human consciousness is seen as a quantum

mechanical phenomenon in that it behaves as both a wave and a particle, in

compliance with the wave-particle duality of quantum mechanics. Even more

importantly, according to Wendt, “consciousness does not end with us –

collective consciousnesses are possible, and particularly a conscious state”

(Keeley 2007: 425). This is exactly what interests Wendt most of all. He

assumes QCH might be true and explores its implications for thinking about

human behaviour, society, state and the international system. However, one

crucial reservation he shares about QCH concerns decoherence, since QCH

under its current formulation cannot explain “how can the brain sustain and

307

aggregate trillions of quantum events into one experience of quantum

subjectivity” (Wendt 2006b).

Furthermore, Wendt (2010) argues for a ‘holographic model’ of the

relationship between individuals and society. He does this in order to avoid the

recasting of intersubjective phenomena in terms of individual choices, in which

supervenience at the end of the day results. Attempting to transcend the

distinction of individual and societies, Wendt takes refuge in the adoption of

the ‘holographic hypothesis.’ How do all these things really ‘apply’ to a new

conceiving of the structure of the international system? The whole idea is

rather simple. Unlike photographs, holograms store all their information in

every part of the image. So, if cut into pieces, each piece will contain a smaller,

intact and, so to say, ‘inclusive’ version of the original image. Wendt thinks

this is also true of the international system and therefore he considers

individuals to be simple points of it, with each of them bearing in its own mind

or having ‘deposited’ in its own memory all the information needed for the

recreation of the whole international system.

An important ramification of the acceptance of the ‘holographic principle’

when analysing the structure of the international system is located in the

observation that “what all this suggests, in short, is a completely flat ontology

for social science, in which state, the international system, and all other

collective intentions are nothing but individuals acting in material contexts”

(Wendt 2010: 295). Individuals are, this way, brought back in, however

without the principle of Methodological Individualism (MI), since the latter is

not ‘compatible’ with the idea of ‘quantum entanglement’ and its function in

producing collective consciousnesses. The ‘level of analysis issue’ has been

308

restated on the basis that “states and other putative corporate actors are not

really actors, but holographic projections by individuals” (Wendt 2010: 301),

produced during the following process:

...Actions taken with reference to the wave function of the international system (for

example, invading Iraq) will typically affect more people at greater distances than

actions taken to reference to local high school board elections. Thus, in contrast to

the vertical imagery of bottom-up/top-down that dominates the current debate, in a

flat ontology it might be better to start by replacing ‘level’ with the more horizontal

imagery of ‘scale’ (cf. Marston et al 2005), and then thinking about relationships

between scales in terms of ‘inside-out’ vs. ‘outside-in’. Neither direction of

influence would be causal (in a classical sense, anyway), because there is no real

reality beyond the individual to do any causing; thus, on this view it would not

make sense to talk about ‘states’ affecting, or being affected by, the international

system…So while in a formal sense I am suggesting we abandon the discourse of

levels, I am not suggesting that we can currently do without concepts like the state

or international system. But these concepts need to be re-conceptualized and used

in a way consistent with the fact that they are only virtual realities, not real ones.

(Wendt 2010: 303).

There are a number of comments that one may offer with regards to Wendt's

aforementioned views:

1. First and foremost, it is important not to forget that the whole project of

using the ‘holographic principle’ to the study of the international system, which

consequently has led, among others, to a revision of the ‘level of analysis

issue’ in IR, is deeply flawed. This is because, as shown previously in this

309

chapter, the context of the ‘holographic principle’ in Physics is quite different

from the one accorded to it by Wendt, with the purpose to ‘adopt’ and ‘apply’

it to his newly developed ‘quantum social science’.

2. Secondly, Keeley puts forward an array of reasonable questions about

consciousnesses of putative agents on different levels of the international

system, which I fully endorse:

If we accept the argument for the state as a high-order consciousness, other

questions inevitably arise. Does this imply that decision processes visible to us are

(on our level) quantum processes of decoherence? Are we to the state as neurons to

the mind? Further, as Wendt seems to recognize, there is no inherent need to stop

the generation of such consciousnesses with the state. In that case the question of

governments arises: why should we accept the state yet reject governments (not to

mention other collectives) as consciousnesses? The number of actors in

international relations, as well as the levels of analysis to be handled, would be

increased significantly. Moreover, it would seem possible for individuals to

participate in more than one such consciousness simultaneously. If such higher-

order consciousnesses exist, how could we know it? Would they know we exist?

What would be the connection between our processes and their thoughts and

activity: could we deliberately influence them? What connection, if any, would

their purposes have with ours? Would we be at best as flies to wanton boys?

Particularly if Wendt wishes to develop his argument in a teleological vein, there

could be serious problems for a progressive element in that argument. (Keeley

2007: 427-428).

310

3. When it comes to the revised ‘level of analysis issue’ which is based on the

‘holographic principle’, I am bound to say that, were it correct, it could have

been of some help to the study of some interesting political phenomena

emerging in the international political landscape such as that of the

Europeanization of national foreign policies of the EU Member States. The

latter cannot be sufficiently explained with the methods provided by the

traditional IR Theory and Comparative Politics but, on the contrary, necessitate

the invention and synthesis of new ideas and concepts.189 Questions of the sort,

“How to conceptualize the process (e.g. is it specific to EU Member States?)”

or “What is changing and what are the mechanisms and direction of change

(top-down from the EU Member States, bottom-up, or socialization)?” (Wong

and Hill 2011b: 4) belong, to a certain extent, to the problematic of the ‘level

of analysis issue’. However, I am afraid that Wendt’s new approach to the

‘level of analysis issue’, though much more ambitious than the previous and

well known ones in the rich bibliography of IR Theory on this topic, makes the

study of the Europeanization of national foreign policies of the EU Member

States much more complicated. This is because it does not provide us with any

new analytic conceptual tools suitable to explain different aspects of this

complex phenomenon. It seems to me that different FPA middle-range theories

that aim to shed light on the different manifestations and epiphenomena of the

aforementioned Europeanization process, along with historical studies of the

routes each one of these states followed in their attempts to reach this goal, can

offer better services in explaining the Europeanization of the national foreign

189 For this line of argument see Mouzakitis (2011).

311

policies of the EU Member States190 than Wendt’s analysis of the international

system, which is based on the idea of collective quantum consciousnesses of

corporate agents and has the individual as its basic unit. For if this very idea of

quantum consciousness indeed proves to be flawed (as I have tried to show that

it is), the whole fabric falls apart immediately. In a more general note, though

grand-theories, such as that of Wendt's, raise great hopes for explaining

complex social and political phenomena, it is a commonplace in IR Theory and

FPA that “attempts at grand generalizations always bring forth a shoal of

objections to single-factor explanations of a complex world” (Hill 2011: 84).

4. The statement that “individuals are back in” makes me wonder whether this

statement amounts to anything more than a mere truism, since Wendt seems to

have nothing more substantial to contribute to the long enduring discussion of

individuals’ role in interpreting or even shaping large-scale political events. It

is my belief that History and Foreign Policy Analysis (FPA) are much more

suitable than Wendt’s ‘holographic hypothesis’ for this goal, in the sense that

they offer concrete personified examples of individuals who act and produce

important political events on micro- and macro- scale as well as sophisticated

methodology, which can be used in the analysis of such phenomena. With

respect to the above, it is worth recalling Christopher Hill's claims that

190 For instance, Hill (1983), Hill (1996a), Wong and Hill (2011a), Manners and Whitman (2000) and Tonra (2001) constitute good examples of the study of the Europeanization of national foreign policies of the EU Member States, with each of these volumes containing both introductions (where methods and goals of the study are presented) and conclusions (in which attempts to categorizing the findings of the papers included in each volume are of considerable help to a comprehensive theorizing of this Europeanization process in different but successive time periods of its development).

312

...Individuals are not simply passive victims of impersonal forces. They are always

agents of some kind and in some context, and deserve attention for their views and

actions as well as their pain. (Hill 1999: 122) 191

And that

While academics tend to discuss IR at the level of the system, once the focus is

shifted to agency it is all too easy to be seduced by the fascination of events, the

charisma of personalities and the demands for what Churchill called ‘action this

day’. But beyond all this is the challenge of acting to develop a framework within

which particular actions may legitimately be pursued and cooperation may have a

chance of holding. (Hill 2011: 97)

It may be much worthier to consider the above acute observations when

studying agency in international politics, than Wendt’s rather simple reminder

of the crucial role individuals play in the development of the international

system. For instance, Wendt does not offer anything more in the way of

explaining what it actually means for someone to act politically as an

individual in the modern complex world, which forms can his actions take and

under which presuppositions and circumstances are these actions feasible,

which are the probabilities for their success, and so on.

One counter argument which might be put forward is that these questions do

not belong to the anxieties and considerations of an IR theorist who aspires to

establish a grand theory of international politics which should be based on the

191 Hill wrote this paper in reference to the war diaries 1939-1945 of Gustave Folcher, a French peasant, on the occasion of his translating them from the French to the English language. Herein he reflects on their importance for the study of History and IR.

313

idea of the existence of both individual and corporate consciousnesses, and

which will be able to explain, by reduction to it, all the political phenomena

that gradually emerge from political actions of single and corporate agents who

act on different but intersecting political levels, domestic, foreign and

international. The questions in focus, the counter argument goes, matter to

those IR scholars who engage in empirical research or who are interested in

middle-range theories. Even so, the essence of the initial question remains:

what can Wendt’s idea about quantum consciousness tell us, at least for the

time being (when, let us not forget, we have not understood all of quantum

mechanics and have understood little on consciousness) which is new about the

importance of the individual acting in the political landscape, national and

international?

5. Furthermore, it seems important to point out that by focusing once more on

ontology (namely, the ontology of the quantum wave-particle duality, this

time) and by avoiding any methodological and epistemological considerations

– by establishing ‘a new non-foundational epistemology’ (Griffiths, Roach and

Solomon 2009: 159) – Wendt sticks to the meta-theoretical framework of his

earlier developed SR (Wendt 1987, 1999). This holds, despite the fact that he

has now moderated his previous concern for (unobservable) social structures

by according priority to the agent (individual) over the structure (meaning the

state, the government, the international system and other corporate units).

Even in the case where one, unlike me, acclaims Wendt’s new ambitious

theory by considering it to be a kind of ‘social physics’ and a part of the

positivist methodologies tradition (see, for instance, Curtis and Koivisto 2010:

314

438-439, as cited earlier in this chapter), there does not seem to be a problem

of incompatibility between the alleged positivist methodology of Wendt’s

quantum social science and SR; the latter being this theory’s meta-theoretical

framework. This is more the case, if we are to adopt Jackson’s definition of ‘a

pluralist science of IR’, which is based on methodological pluralism (not

methodological relativism) and “poses the challenge of dealing with bodies of

warranted knowledge stemming from philosophically incompatible

methodologies” (Jackson 2011: 210).192

6. Another interesting comment to be made on Wendt’s ‘quantum

consciousness theory’ is that it implies that “the more mental causality and thus

autonomy of ‘consciousness’ is granted, the more an ontologically based

constructivism becomes implausible” (Kessler 2007: 243). Kessler also argues

that in the event one can trace affinities between Wendt’s turn to quantum and

other IR theories, he can’t help tracking such resemblances in radical versions

of IR theories which either are exclusively Wittgensteinian in their approaches

or follow, more generally, the so-called ‘linguistic turn’. This happens, he

claims, because theories of this sort are interested in speaking as an activity,

i.e. showing the indeterminacy of world and inter-subjective objectivity of facts

192 The obvious question of course, is how this is feasible. According to Jackson,

Because there is no methodological neutral metalanguage into which we could reliably translate our warranted knowledge-claims and have them be globally understood, methodological pluralism sets up a variety of contentious conversations and efforts to appreciate the insights of alternative ways of producing knowledge while avoiding the temptation to universalize our own modes of conducting scientific inquiry. (Jackson 2011: 210)

Although I do not find the meaning of this passage perfectly clear, especially regarding the ways and ‘methods’ that these conversations may lead to the described desired goal, what I derive from it is the claim that SR is compatible with the methodological positivism of Wendt’s quantum social science; a claim with which I do agree.

315

“which leads, according to Stefano Guzzini, to a micro and macro feedback

loop of description and action” (Kessler 2007: 266). Kessler presumably sees

affinities between ‘the indeterminacy of world’ and the ‘uncertainty principle’

of quantum mechanics, on the one hand, and the ‘inter-subjective objectivity of

facts’ and the ‘quantum measurement problem’, on the other. However, as I

have explained before, this is not the case, and as such, Kessler’s argument is

rendered faulty.

7. At the beginning of this chapter I have quoted a statement made by Griffiths,

Roach and Solomon, according to which, the primary aim of Wendt’s quantum

social science project “is to suggest the possibilities of what he calls ‘capacity

for collective self-consciousness’” (Griffiths, Roach and Solomon 2009: 158-

159).

However, as I have repeatedly noted, I seriously doubt whether one should

take Wendt’s quantum theory of consciousness and its implications for the

formation of collective self-consciousness at their face value, at least for the

time being and given that the theory is at its infantile stages of development.

Moreover, one should also not take Wendt’s quantum theory of consciousness

and its implications for the formation of collective self-consciousness at their

face value, because that would be done at the expense of relevant insights,

which bear on collective self-consciousness and are drawn on cognitive

psychology and psychoanalysis (due to the latter’s attempts to reach the

subconscious).

It is my belief that the above point is very important, because collective self-

consciousness is crucial when studying the formation of collective identity, a

316

major issue of Wendt’s first influential book (Wendt 1999). I discussed this at

length in Chapter 4 of the thesis, and as such, nothing more will be added here.

Collective self-consciousness is also essential, for the analysis of foreign

policy decision-making, especially when one comes to the theory of

groupthink, where collective psychology provides us with insightful

observations on human behaviour. In his discussion of the psychological factor

in the framework of foreign policy decision-making analysis, after calling our

attention to the commonsensical but, at the same time, important assumption

that “there is a difference, however uncertain, between decision-makers’

psychological environment (how they perceive the world) and their operational

environment (events as they happen independent of one person’s perception)”

(Hill 2003: 111, italics in the original), Christopher Hill then relies upon it and

goes on to examine, among others, the role of rationality and irrationality in

the groupthink. Suffice it to cite here only his concluding words on this issue:

Decision-makers at times have no choice but to accept the limits on rationality and

bring other qualities into play, some of them using the emotional rather than the

calculating side of the brain. Miriam Steiner has argued, in a brilliant and still

unsurpassed article, that ‘in a world with important nonrationalistic elements, true

rationality requires that nonrationalistic capabilities and skills be appreciated and

developed side by side with the rationalistic ones’ (Steiner 1983: 413) Feeling and

intuition are just as vital attributes of decision-makers as thinking and sense-based

observation. Thus the overwhelming mass of impediments to rationality are not

quite the obstacles they seem; some, indeed, cry out for a-rational, if not non-

rational, techniques, and concepts like judgement, leadership, empathy and

charisma always imply at the very least a combination of the intellectual and

317

emotional sides of personality. Ultimately, both cognitive psychology and

psychoanalysis have done a lot to contribute to our understanding of how foreign

policy decisions are and should be made. (Hill 1999: 116)

It is true that Karl Popper in his classic, The Poverty of Historicism, argues that

“the social sciences are comparatively independent of psychological

assumptions, and that psychology can be treated, not as the basis of all social

sciences, but as one social science among others” (Popper [1957] 1997: 142).

However, I cannot help but notice that although psychology should probably

not be treated as the basis of all social sciences, cognitive psychology has been

such a substantial contribution to the understanding of foreign policy decision-

making that it should not be ignored.

8. Finally, on a more general point, in his attempt to defend the methodological

pluralism of the English School (ES), Richard Little argues that, albeit

tentatively, “some of the more surprising implications of the ES approach to

methodology have started to be teased out in Alexander Wendt’s most recent

work” (Little 2008: 100). However, he goes on, “few if any ES theorists…are

likely to engage with Wendt’s quantum move. They are much more likely to be

sympathetic with the view that epistemological and ontological problems will

be resolved in the hurly-burly of doing research” (ibid: 94).

Little seems convinced that Wendt’s quantum social science project signals a

turning point in the methodological study of social sciences, because of the

abandonment of the Cartesian dualism which it entails:

318

But in essence, sub-atomic phenomena can be described in two irreducible and

non-equivalent ways – as either waves or particles and, as a consequence it is not

possible in principle, to know both the position and momentum of a particle at the

same time. It follows that the idea of unitary and integrated knowledge is

inherently impossible, because we necessarily require conflicting narratives to

describe sub-atomic phenomena. Moreover, in providing these narratives, it also

has to be accepted that the subject-object distinction breaks down. Quantum

physicists, therefore, have opened up a Pandora’s Box to reveal, among other

things, a post-modern world of alternative realities where the conception of

causality breaks down. (ibid: 93)

If one bears in mind my comments on and objections to Wendt’s understanding

of QM, which I have deployed earlier in this chapter, one can come to

understand why I disagree with the Little’s conclusions. Given that the

description of sub-atomic phenomena does not involve conflicting narratives

and the conception of causality does not break down in QM (it becomes

probabilistic), I cannot conceive what is meant by “a post-modern world of

alternative realities where the conception of causality breaks down”.

Furthermore, Little holds that the most important methodological

implication of Wendt’s quantum social science project is that

…it has led him [Wendt] to shift his position on methodological pluralism and to

reject the attempt to achieve a synthesis of positivism and interpretivism. Instead,

he argues that quantum naturalism reaffirms the position that explanations sought

by positivists and the understanding pursued by interpretivists are both necessary

and complementary ways of viewing the social world. (ibid: 94)

319

Be that as it may, what is of real interest to me is that Little argues that the

above conclusion “might not come of a surprise to the ES” (ibid: 94), because

“the founders were, in essence, intuitive or perhaps even unintentional

methodological pluralists” (ibid: 94). Although I cannot understand how one

can unintentionally subscribe to methodological pluralism, let alone

unintentionally ‘serve’ it in his theoretical or empirical research, I do agree

with Little’s point that “the complexity and diversity of the ontological and

epistemological assumptions that underpin the ES approach to theory and

history have come about, to some extent, as a consequence of not paying close

attention to ontological and epistemological questions” (ibid: 94, emphasis

added). I think that this observation is important since one of the main

problematiques of this thesis, as discussed in chapter 4, deals exactly with the

question of the extent to which second-order IR meta-theoretical considerations

determine the ontological and methodological plurality and sophistication of

the first-order IR theories that they are taken to underpin.

Along the same line of argument, Little contends that the methodological

pluralism which is implied by Wendt’s quantum social science project can also

be applied to many working historians (and here one must take into account the

impact historians have had on the English School’s research agenda):

Historians almost invariably acknowledge that there are always at least two stories

to tell, one from the inside and the other from the outside. They know that they

operate from a potentially privileged position because they can endeavour to

recover the viewpoint of the historical actors, but also have the luxury of knowing

what happened after the event. So, self-fulfilling prophecies and unanticipated

consequences, for example, can become crucial elements of any historical

320

narrative. Indeed, historians necessarily have to make sense of events in the light of

subsequent events. What is distinctive about the ES is not simply the focus on

structural constraints, but also the recognition that the impact of structural

constraints have to be examined in the light of both the separation and the

interaction between facts and values, the relationship between the story told from

the outside and the inside, the fact that the actors have a conception of both the past

and the future, and the link between actors and analysts. These complexities map

quite closely onto Wendt’s quantum perspective on methodological pluralism.

(ibid: 94)

With regards to the above, I really doubt if we needed “Wendt’s quantum

perspective on methodological pluralism” in order to conceive the above

complexities, which definitely characterize the English School’s

methodological apparatus. These can be found out directly if one reads

carefully the existing bibliography on the English School. Now, if one wants to

come to grips with such difficult questions as “What is History?” and “What

historians actually do?”, one can refer to a number of great works which

demonstrate utmost originality (for instance, Carr ([1961] 2001), Collingwood

([1946] 1973)) or strong analytical power (for instance, Evans (1997).

Moreover, on the Philosophy of History (which deals with these and other

similar questions), one can use Dray (1964) and Walsh ([1977] 2008). As a

consequence, one does not need to resort, as Little wrongly suggests, to

Wendt’s new theory in order to shed light on the above issues.

321

5.6 Conclusion

The foregoing considerations articulate several lines of criticism of Wendt’s

(2006a, 2010) preliminary sketch of a road towards a quantum social science.

Some pertain to issues of methodology while others to issues of content. As to

methodology, I argued that Wendt fails to offer precise arguments in support of

his claims. Moreover, the arguments the reader can distill from Wendt’s text

often fail to be cogent due to violations of well-known criteria that should

govern the deployment of the pertinent argument forms. Coming to the main

argument of his intellectual venture, which concerns the literal application of

quantum mechanics to the study of social science, I argued that Wendt

misconceives the context and function of many basic concepts of quantum

physics. Moreover, while he also misconceives the context and function of

their philosophical interpretations and ramifications, he remains – whether

knowingly or not – too eclectic as to the metaphysical assumptions of his

intellectual enterprise. Thus, despite his intentions, the results he produces for

IR Theory are of minor importance as they do not constitute any scientific

progress. Of course, it should be granted that some inaccuracy and vagueness

may be due to the fact that Wendt's project is at its infantile stages of

development. But any research project, no matter how ambitious or how early

on in its development stage it is, must have a clear core and orientation.

Otherwise, there are neither heuristic guidelines for the advancement of the

programme nor criteria that will distinguish success from failure.

Overall, Wendt’s quantum programme shares a feature, which is common to

many attempts that appeal to quantum physics in order to ‘solve’ some thorny

philosophical problem and is plainly revealed in Wendt’s own words:

322

In short, the nature of the quantum is no less a mystery than consciousness…. The

quantum consciousness hypothesis suggests that the two mysteries have a common

solution, namely that the quantum in effect is consciousness, which in some form

goes all the way down in matter. (Wendt 2006a: 190)

But no mystery, as long as it remains a mystery, can solve another mystery.

And the idea that progress will be achieved by reducing the total number of

mysteries to be resolved amounts, at this level of generality, to mere wishful

thinking. True, the strange but fascinating facets of quantum reality – in

particular, the miracle of quantum measurement – have been marshaled to

resolve several baffling puzzles. But the trick that makes such attempts

appealing is simple: from a miracle almost anything follows!

323

6

CONCLUSION

6.1 The Contributions

My aim in this section is to outline some of the contributions that the argument

in the present thesis offers in terms of the critique of the introduction and use of

SR and CR in IR as well as in terms of other issues which are discussed in the

thesis (e.g. SR in the philosophy of science, Bhaskar’s CR) and pertain to its

problematique.

In Chapter 2, I examined SR as an autonomous pursuit in the philosophy of

science and found that the outcome of the realism/anti-realism debate has been

that one should no longer go uncritically for an adoption of the standard global

conception of SR. This is of utmost relevance, since one of the central

arguments I make in the thesis is that the manner in which SR has been

imported and used in IR meta-theory does not appreciate adequately the grave

theoretical implications that SR has for any strong, global realist construal of

IR. The standard brand of SR adopted by IR meta-theory is considered to be the

definitive and conclusive version of philosophical realism; this is mistaken,

since many issues on SR remain unresolved, whereas different interpretations

of realism abound. Picking the strongest version among them rather uncritically

as the definite and final interpretation of realism and importing it into IR meta-

theory is a risky business. I also noted how the adoption of an unreal,

‘mythological’ conception of science is uncritical and permeates much of the

work in IR meta-theory – notably Alexander Wendt’s and Colin Wight’s.

324

Consequently, the aforementioned works inherit all the problems that go with

such an adoption.

In Chapter 3, I embarked upon a very close reading of Bhaskar’s main

statement of his particular version of SR, also known as CR. The chapter is of a

thoroughly philosophical character, focusing solely on the examination and

structuration of a series of philosophical arguments. Such a detailed

examination of CR is necessary, because it is this Bhaskarian version of SR,

which constitutes the hegemonic meta-theoretical discourse in the SR debate

within IR. Failure to do such an examination can lead one to an unreflective

stance towards SR’s and philosophy of science’s place in general in IR. The

main original contributions to the understanding of Bhaskar’s philosophy

offered in this chapter could be summarized as follows:

1) I offer an argument that explains in what sense one can understand

Bhaskar’s notion of a ‘transitive object of science’.

2) I elucidate Bhaskar’s distinction of the Humean and the Kantian versions

of ‘classical empiricism’, while explaining, through an original argument

based on the notions of subjectivity, experienciability and the ‘empirical

world’, why both versions are committed to a common ontology, which

Bhaskar terms ‘empirical realism’.

3) I dissolve a very common misunderstanding, one that appears much too

often in the meta-theoretical debates of IR, concerning the nature of a

transcendental argument. This allows me to claim that Bhaskar makes

constant use of ‘transcendental argumentation’ in order to demonstrate the

intransitive dimension of scientific practice.

325

4) I place Bhaskar’s philosophy in the context of the history of philosophy,

particularly in reference to Berkeley, Hume and Kant as well as Aristotle.

5) I present a novel codification of his first transcendental argument based on

the notion of sense-perception.

6) I present an original codification of his second and most famous

transcendental argument based on the notion of experimental activity.

7) I explain in a novel way how the notions of structure, generative

mechanism and natural tendency of a thing are intertwined.

8) I differentiate in an explicit way three levels of analysis: philosophy of

science, methodology and ontology. I make several original claims about

Bhaskar’s use of the notions of ‘ontology’ and ‘transcendental’.

9) I present a critique of Bhaskar’s implicit understanding of Kant’s theory of

causation.

10) I present the novel statement that Bhaskar makes use of Kantian

methodology in order to arrive to an utterly un-Kantian and thoroughly, yet

inexplicitly, Aristotelian ontology.

11) I explain in a brief, yet hopefully clear, manner how Bhaskar’s ontology of

generative mechanisms is to be associated with Aristotelian ontology of

substances.

12) I present an explanation of the appeal of Bhaskar’s philosophy of science

to working social scientists, as Bhaskar’s philosophy presents a richer

picture of the subject-matter of social science, which goes beyond

observable patterns of events.

326

In Chapter 4, I argued that Wendt has imported the strongest version of SR into

IR, a ‘move’ which is risky since many issues on SR remain unresolved and

different interpretations of realism abound (as I have shown in Chapter 2).

Unfortunately, those IR scholars who have got involved in the SR debate in IR

do not seem be aware of this ‘complex situation’, at the expense of the quality

of the argumentation about SR and CR within IR. I have also shown that many

IR scholars who have come to grips with these issues have not been able to

distinguish between SR and CR, the Bhaskarian version of SR, which has been

introduced by Wight and other critical realists into IR and which has dominated

the field. The illumination of this distinction is of some importance – without it

being, however, novel – since Wendt and Wight have based their meta-theories

on different premises and this is not inconsequential. For instance, although

they both prioritize ontology over epistemology, Wight argues that CR is

incompatible with positivism, whereas Wendt proposes a methodological

‘reconciliation’ between positivist and post-positivist approaches to the study of

IR. Therefore, a second small contribution I have made in the framework of the

Chapter in focus is that I have highlighted the differing implications CR and SR

have not only for meta-theory (see the above example) but also for substantive

IR theory (for instance, the adoption of the notion of ‘constitutive explanation’

helps us search, on both the theoretical and empirical research levels, for causes

in places which are sidelined by the ‘causal explanation’, which is mainly used

by positivist IR theories, e.g. neorealism). Furthermore, I have emphasized that

IR critical realists argue that meta-theoretical choices are ‘politically charged’

and I agreed with Chris Brown that CR may contribute to the resurgence of

historical materialism studies in IR.

327

In Chapter 5, I came up with a critique of Wendt’s ‘quantum social science’

project, which is also based on SR as its meta-theoretical background. There are

very few critiques of this project and this explains why this Chapter is novel

with regards to most of its content. I have criticized Wendt’s ‘quantum social

science’ with regard to its methodology, understanding of QM, metaphysics,

and application to IR. My comments on Wendt’s understanding of QM

illustrate how Wendt uses QM literally and not metaphorically. Furthermore,

nowhere in the IR literature is to be found any critical approach to Roger

Penrose’s ‘quantum consciousness theory’, as it is mainly employed in his two

successive and widely discussed books, The Emperor’s New Mind, and

Shadows of the Mind. Consequently, I discussed Penrose’s work on

consciousness, as it has been presented in books and academic journals of

Philosophy, Physics, Mathematics, Logic, Artificial Intelligence and Biology,

in order to show that, despite its sophistication, his theory of consciousness is

highly contested. This is crucial because it makes Wendt’s ‘quantum

consciousness hypothesis’ problematic, given that the latter relies heavily on

Penrose’s hypothesis. If one takes on board that this hypothesis lies in the

foundations of Wendt’s ‘quantum social science’ project, one understands that

in the eventuality that the hypothesis is erroneous, Wendt’s theory falls apart.

Needless to say that, as a consequence, Wendt’s attempt to hinge on ‘quantum

collective consciousnesses’ of states and the international system itself in order

to explain international politics fails, too.

328

6.2 Directions for further research

6.2.1 The agent-structure problem in a nutshell

As I argued in this thesis, Wendt was led to SR due to his effort to resolve the

‘agent-structure problem’ in IR Theory. In his attempt to grant causal powers to

what he terms structure, e.g. something which is unobservable but can,

nonetheless, be inferred by its effects, he chose SR as the meta-theoretical

background of his work. He did this for the simple reason that one of SR’s three

basic assumptions is that the world consists of both observable and

unobservable entities. However, in both his 1987 paper on the agent-structure

problem in IR and his STIP, Wendt fails to come up with any new theoretical

version of the agent-structure model. Moreover, he is absolutely explicit as to

the absence of any intention on his part to get involved in any empirical work,

which could bring to the fore concrete examples of the relationship between

real-worldly agents and structures. He adopts an agent-structure dualism

scheme, that is, he endorses the view according to which agent and structure are

two distinct entities, which affect greatly one another. This happens because he

takes on board Margaret Archer’s social-theoretical work on the agent-structure

issue (Archer 1982, 1995), which is based on Bhaskar’s CR and therefore

admits the aforementioned dualism scheme with respect to the enduring

relationship between agent and structure. In other words, although Wendt

includes Antony Giddens’ considerations about the agent-structure problem in

his own examination of the same problem, he negates Giddens’ view of the

agent-structure duality (Giddens 1979, 1984). The latter involves seeing agent

and structure as two sides of the same coin, meaning that each one of these

entities is absolutely dependent on the other and shapes it in a decisive manner;

329

according to Giddens, the social structure is both the medium and the outcome

of social action.

Colin Wight is another thinker who relies heavily on Archer’s and Bhaskar’s

CR. Thus, while, in general, he is aligned with Wendt on the agent-structure

issue (meaning that he also endorses agent-structure dualism), Wight, unlike

Wendt, offers a typology of structures (as I have shown in Chapter 4),

something which Wendt does not do. The main difference between the two

thinkers, however, lies in the fact that Wendt is more selective than Wight in

his choice of the theoretical model through which he addresses the agent-

structure issue. Whereas Wight follows stricto sensu the Bhaskarian ontology,

methodology and epistemology, Wendt’s SR draws on ideas and notions from a

richer bibliography on SR, sociological theory and, more recently, science.

Moreover, it does this despite the fact that it remains faithful to Bhaskar’s CR

regarding the choice of its basic ontological assumptions. Heikki Patomäki

(2002), Milja Kurki (2007, 2008), David Dessler (1989) and others who also

endorse Archer’s and Bhaskar’s CR, do not distance themselves from the

dominant explanation scheme regarding the relationship between agent and

structure, which they assume to be that of a dualism.

When examining the agent-structure issue with respect to the existing FPA

and European Foreign Policy Analysis (henceforth EFPA), one observes that

the theoretical reflections on it are poorer in FPA and EFPA than they are in IR

Theory. Nevertheless, one can still find interesting reflections on this issue in

Carlsnaes (1992), Tonra (2001), Knudsen (1994), and Wœver (1994).

Furthermore, the majority of FPA scholars quote Giddens (1979, 1984) rather

uncritically when they refer – usually in a vague manner – to actors and

330

structures; a few of them, with Walter Carlsnaes being the most eminent, rely

on the works of Dessler (1989), Archer (1982, 1995, 2000, 2003) and Bhaskar

([1975] 2008, 1979), in order to provide us with their own model of studying it.

Indeed, Carlsnaes conceived his own model of the agent-structure relationship

which he demonstrates in every FPA paper he writes on this subject (Carlsnaes

1992, 1994, 2002, 2006).

My exposition above is a very brief, albeit neither inaccurate nor misleading,

presentation of the agent-structure problematique in the FPA and EFPA

literature.

6.2.2 A new approach to the agent-structure problem

It is my belief that a further original contribution to the discussion of the agent-

structure problematique could result from a focus which combines sociological

theory, IR Theory and FPA. IR is a social science and as such it could be

privileged from its contact with and learning from other social sciences and

sociological theory. It is also true that the philosophy of science and the

philosophy of social sciences are important to the study of the foundations of

natural and social sciences, respectively. However, it is a mistake to substitute

philosophy of science for sociological theory when dealing with sociological

issues. In the same vein, it is implausible to substitute philosophy of science for

IR Theory, International Political Theory, or FPA, when studying the agent-

structure problem at the levels of the national and the international.

The detailed discussion of the agent-structure issue within the framework of

sociological theory may help IR scholars see it in its complexity. With the

exception of Giddens, who has coined the agent-structure duality explanatory

331

scheme, there are many other distinguished sociologists who have dealt with

the same problem at length. One example is Nicos Mouzelis (2008) who draws

on Giddens, Parsons, Habermas, Alexander and Lockwood, to come up with a

more refined explanation of the complex relationship between agent and

structure. He argues that, while at times the relationship between agent and

structures takes the form of a duality, at some other times it is the dualism that

characterizes that relationship. This depends on many factors, one of which is

the given location of the agents and the structures in the micro- and macro-

scale of the social and political environment. This distinction between duality

and dualism with respect to the relationship between agent and structure, and

the observation that this relationship sometimes takes the shape of dualism,

while other times takes the shape of duality (depending on the location of the

actor vis-à-vis that of the structure), cannot be found in current IR theories

about the agent-structure issue and this may make the difference in future

works on the same topic.

Furthermore, one takes the risk of getting lost in endless and rather sterile

meta-theoretical discussions about the agent-structure issue, if one loses sight of

the kinds of agents and structures one can locate in IR, let alone that some of

them may not remain unchangeable or continue existing in the long durée; as

Christopher Hill notes, one should, instead, confront the “‘first order’ questions

of what entities, actors, units to admit of in world politics” (Hill 2003: 313).

One must investigate who the major players (actors) in international politics

are, how important each of them is, and, finally, under which conditions

political action can occur” (Hill 2011). Only then can one estimate if change as

a result of political action (agency) is possible, how much change is reasonable

332

to expect, and from whom. In other words, it is crucial that one should be able

to make concrete empirical research proposals to test his IR theoretical model

of agents and structures and their relationship. I would only like to reiterate that

the lack of such proposals is probably the most serious shortcoming of the

Wendt’s and Wight’s models in IR Theory.

More generally, in trying to do justice to the relevant meta-theoretical

considerations about the agent-structure issue in IR and FPA, Hill argues that

In one sense the agency-structure debate has been good for foreign policy analysis.

It has returned the perennial issues of causation, freedom and determinism to the

agenda of International Relations and it has led to a sharper examination of the

rather unsophisticated conceptual basis of some foreign policy studies. On the other

hand, it has often been presented as the agency-structure ‘problem’, which by

extension should admit a ‘solution’. This is not the approach taken here, where the

mathematical analogy is seen as inappropriate to the immense political and

historical complexities facing all those who wish to understand foreign policy.

Rather it is assumed that causation always involves both structures and agencies,

and that – as a number of authors have pointed out, following Antony Giddens –

the two kinds of phenomena help to constitute each other in a perpetual process of

interaction. This means, by definition, that it is impossible to come to fixed

conclusions about the limits to agents’ freedom of choice or their capacity for

impact (two different things). We may analyse the parameters of choice, constraint

and change but human beings will always have the ‘wiggle room’ of specific

historical circumstances in which to remake at least some of the world. (Hill 2003:

26)

333

Given the above, there is great scope of elaboration on the agent-structure

problematique in the areas of FPA and EFPA, by, first, using tools of

sociological middle-range theories which can account for the observed effects

of the relationship between various kinds of observable agents and

unobservable structures and, second, by taking on board the specific

characteristics of FPA and EFPA (White (2001, 2004a, 2004b), Bretherton &

Vogler (2006), Tonra & Christensen (2004), Carlsnaes & Smith (1994), Tonra

(2001), Keukeleire & MacNaughtan (2008), Hill (2003), Hill & Smith (2011a),

Hill & Smith (2011b), Jørgensen (2004), etc.). For instance, one such tool that

may be adopted from sociological middle-range theories is the ‘strategic-

relational approach’, which has been developed by Bob Jessop within the

framework of CR (Jessop (2002) and Colin Hay (2002) – see also the debate on

the agent-structure issue between Hay (2005) and McAnulla (2005)). The

‘strategic-relational approach’ has been used as a theoretical tool of FPA,

especially when one wants to study how actors meet their environment (which

seems to be a reformulation of ‘the agent-structure problem’ in the FPA

language); for instance, see Brighi 2005: 79-83, and Brighi & Hill 2008: 119-

122. Furthermore, it has been used by critical realists, like Marjo Koivisto, who

relies on the concept of structuration and the ‘strategic relational position’, in

order to account for the state’s normative capacities in IR (Koivisto 2012: 46-

8).

However, irrespective of how sophisticated the models I am going to use

may be, what I will try to bear in mind and never forget is the existence of the

‘wiggle room’ of which Hill has spoken in the above cited passage. This

334

pragmatic expression in fact denotes the fundamental freedom which human

beings always retain to make their own history.

335

7

BIBLIOGRAPHY

Aaronson, S. (2013) Quantum Computing Since Democritus. Cambridge:

Cambridge University Press.

Achinstein, P. (2002) “Is there a Valid Experimental Argument for Scientific

Realism?”, Journal of Philosophy, 99(9), 470-95.

Adkins, A.W.H. (1978) “Theoria versus Praxis in the Nicomachean Ethics and

the Republic”, Classical Philology, 73(4), 297-313.

Adorno, Theodor W. et al. (1976) The Positivist Dispute in German Sociology.

London: Heinemann.

Agassi, J. (1975) Science in Flux. Dordrecht: Reidel.

Aggestam, L. & Hill, C. (2008) “The challenge of multiculturalism in European

foreign policy”, International Affairs, 84(1), 97-114.

Ainley, K. (2006) Rethinking Agency & Responsibility in Contemporary

International Political Theory, London School of Economics, PhD thesis

(unpublished).

Akrivoulis, D.E. (2002) The Quantum Politics Metaphor in International

Politics: Revising American Newtonianism, University of Kent at

Canterbury, PhD thesis (unpublished).

Akrivoulis, D.E. (2007) “The ‘Quantum Politics’ Metaphor in International

Relations: Towards a Hermeneutics of Political Metaphoricity”, available at:

http://www.psa.ac.uk/journals/pdf/5/2007/Akrivoulis.pdf (Paper presented

336

at the 57th Political Studies Association Annual Conference, University of

Bath, UK, 11-13 April 2007).

Albert, D.Z. (1992) Quantum Mechanics and Experience. Cambridge, MA:

Harvard University Press.

Albert, M. et al. (eds.) (2010) New Systems Theories of World Politics. New

York: Palgrave Macmillan.

Alker, H. (2000) “On Learning from Wendt”, Review of International Studies,

26(1), 141-50.

Alspector-Kelly, M. (2001), “Should the Empiricist be a Constructive

Empiricist?”, Philosophy of Science, 68(4), 413-31.

Alspector-Kelly, M. (2004) “Seeing the Unobservable: van Fraassen on the

Limits of Experience”, Synthese, 140(3), 331-53.

Anscombe, G.E.M. & Geach, P.T. (1961) Three Philosophers. Oxford: Basil

Blackwell.

Archer, M. (1982) “Morphogenesis versus structuration: On combining

structure and action”, British Journal of Sociology, 33(4), 455-83.

Archer, M. (1995) Realist Social Theory: the Morphogenetic Approach.

Cambridge: Cambridge University Press.

Archer, M. (2000) Being Human: the Problem of Agency. Cambridge:

Cambridge University Press.

Archer, M. (2003) Structure, Agency and the Internal Conversation.

Cambridge: Cambridge University Press.

Arendt, H. (1999) The Human Condition. Chicago: University of Chicago

Press.

337

Arfi, B. (2012) “Khôra as the condition of possibility of the ontological without

ontology”, Review of International Studies, 38(1), 191-207.

Aspect, A., Grangier, P. & Roger, G. (1982) “Experimental Realization of

Einstein-Podolsky-Rosen-Bohm Gedankenexperiment: A New Violation of

Bell’s Inequalities”, Physical Review Letters, 49(2), 91-94.

Asquith, P.D. & Giere, R.N. (eds.) (1981) PSA 1980, Vol. 2. East Lansing, MI:

Philosophy of Science Association.

Atmanspacher, H. (2006) “Quantum Approaches to Consciousness”, in E.

Zalta (ed.), Stanford Encyclopedia of Philosophy, available online at

http://plato.stanford.edu .

Baars, B.J. (1994) “Roger Penrose and the quest for the quantum soul”,

Journal of Consciousness Studies, 1(2), 261-263.

Baars, B.J. (1995) “Can Physics provide a theory of consciousness?” – Book

Review [of Penrose 1994a], PSYCHE – An Interdisciplinary Journal of

Research on Consciousness, available online at

http://www.theassc.org/files/assc/2330.pdf .

Baaz, M. et al. (eds.) (2011) Kurt Gödel and the Foundations of Mathematics.

Cambridge: Cambridge University Press.

Balashov, Y. & Rosenberg, A. (eds.) (2002) Philosophy of Science –

Contemporary Readings. London: Routledge.

Barkin, J.S. (2010) Realist constructivism: rethinking International Relations

theory. Cambridge: Cambridge University Press.

Barr, M. (1990) Book Review [of Penrose 1989], The American Mathematical

Monthly, 97(10), 938-942.

338

Bhaskar, R. ([1975] 2008) A Realist Theory of Science. London, New York:

Verso.

Bhaskar, R. (1979) The Possibility of Naturalism: A Philosophical critique of

Contemporary Human Sciences. Sussex: Harvester.

Bhaskar, R. (1993) Dialectic: the Pulse of Freedom. London: Verso.

Bhaskar, R. (1994) Plato etc.: the Problems of Philosophy and their Resolution.

London: Verso.

Bhaskar, R. (2003) “Ontology”, in W. Outhwaite (ed.) The Blackwell

Dictionary of Modern Social Thought, 2nd edition. Oxford: Blackwell.

Belot, G. (1996) “Why General Relativity Does Need an Interpretation”,

Philosophy of Science, 63 (Supplement), S80-S88.

Belot, G. (2011) Geometric Possibility. Oxford: Oxford University Press.

Bergström, L. (1984) “Underdetermination and Realism”, Erkenntnis, 21, 349-

65.

Bernstein, S., Lebow, R.N., Stein, J. G. & Weber, S. (2000), “God gave

Physics the Easy Problems: Adapting Social Science to an Unpredictable

World”, European Journal of International Studies, 6(1), 43-76.

Berto, F. (2009) There’s Something About Gödel – The Complete Guide to the

Incompleteness Theorem. West Sussex: Wiley-Blackwell.

Bishop, M.A. (2003) “The Pessimistic Induction, the Flight to Reference and

the Metaphysical Zoo”, International Studies in the Philosophy of Science,

17(2), 161-78.

Bishop, M.A. & Stich S.P. (1998) “The Flight to Reference, or How Not to

Make Progress in the Philosophy of Science”, Philosophy of Science, 65(1),

33-49.

339

Bjola, C. (2007) “Ontology, All the Way Down”: Book Review [of Wight

2006], International Studies Review, 9, 316-18.

Blackburn, S. (1994) The Oxford Dictionary of Philosophy. Oxford: Oxford

University Press.

Blackburn, S. (2002) “Realism: Deconstructing the Debate”, Ratio (new series),

15(2), 111-33.

Bohm, D. ([1980] 1997) Wholeness and the Implicate Order. London:

Routledge.

Bohman, J. (2009) “What is to be done? The science question in International

Relations”, International Theory, 1(3), 488-98.

Bohr, N. ([1949] 1983) “Discussion with Einstein on Epistemological Problems

in Atomic Physics”, in J.A. Wheeler & W.H. Zurek (eds.) Quantum Theory

and Measurement. Princeton, NJ: Princeton University Press.

Booth, K. & Smith, S. (eds.) (1995) International Relations Theory Today.

Cambridge: Polity Press.

Boyd, R. (1973) “Realism, Underdetermination and a Causal Theory of

Evidence”, Nous, 7(1), 1-12.

Boyd, R. (1981) “Scientific Realism and Naturalistic Epistemology”, in P.D.

Asquith & R.N. Giere (eds.) PSA 1980, vol. 2. East Lansing, MI: Philosophy of

Science Association.

Boyd, R. (1983a) “On the Current Status of the Issue of Scientific Realism”,

Erkenntnis, 19, 45-90.

Boyd, R. (1983b) “The Current Status of Scientific Realism” in J. Leplin (ed.)

Scientific Realism. Berkeley: University of California Press.

340

Boyd, R. (1985) “Lex Orandi est Lex Credendi”, in P.M. Churchland & C.A.

Hooker (eds.) Images of Science: Essays on Realism and Empiricism, with a

Reply from Bas C. van Fraassen. Chicago: University of Chicago Press.

Boyd, R (1990) “Realism, Approximate Truth, and Philosophical Method”, in W.

Savage (ed.) Minnesota Sttudies in the Philosophy of Science, vol. 14:

Scientific Theories. Minneapolis: University of Minnesota Press.

Boyd, R. (1992) “Constructivism, Realism, and Philosophical Method”, in J.

Earman (ed.) Inference, Explanation, and Other Frustrations. Berkeley, Los

Angeles: University of California Press.

Boyd, R. (2002) “Scientific Realism”, The Stanford Encyclopedia of Philosophy,

Summer 2006 edition, available online at

http://plato.stanford.edu/entries/scientific-realism/ .

Bretherton, C. & Vogler, J. (2006) The European Union as a Global Actor.

London, New York: Routledge.

Brglez, Milan (2001) “Reconsidering Wendt’s Meta-theory: Blending Scientific

Realism with Social Constructivism”, Journal of International Relations and

Development, 1(4), 339-62.

Brighi, E. (2005) Foreign Policy and the international/domestic nexus: the case

of Italy, University of London - The LSE Department of International

Relations, PhD thesis (unpublished).

Brighi, E. & Hill, C. (2008) “Implementation and behaviour”, in S. Smith, A.

Hadfield, & T. Dunne (eds.) Foreign Policy - Theories, Actors, Cases. Oxford:

Oxford University Press.

341

Bringsjord, S. & Xiao, H. (2000) “A Refutation of Penrose’s Gödelian Case

Against Artificial Intelligence”, available online at

http://homepages.rpi.edu/~faheyj2/SB/SELPAP/PENROSE/pen.sel8.pdf .

Brown, A., Fleetwood, S. & Roberts, J.M. (eds.) (2001) Critical Realism and

Marxism. London: Routledge.

Brown, B. (1990) “How to be Realistic about Inconsistency in Science”,

Studies in the History and Philosophy of Science, 21(2), 281-94.

Brown, C. (1992) International Relations Theory – New Normative

Approaches. New York: Columbia University Press.

Brown, C. (2002) Sovereignty, Rights and Justice – International Political

Theory Today. Cambridge: Polity Press.

Brown, C. (2007) “Situating Critical Realism”, Milllennium: Journal of

International Studies, 35(2), 409-416.

Brown, C. (2010) Practical Judgement in International Political Theory –

Selected Essays. London, New York: Routledge.

Brown, C. (2012) “Review article – Realism: rational or reasonable?”,

International Affairs, 88(4), 857-866.

Brown, C. & Ainley, K. (2009) Understanding International Relations, 4th

edition. New York: Palgrave Macmillan.

Brown, J. (1994) “Tell me where consciousness is bred” – Book Review [of

Penrose 1994], New Scientist, November 19th: 46-47.

Bueno, O. (1997) “Empirical Adequacy: A Partial Structures Approach”,

Studies in the History and Philosophy of Science, 28(4), 585-610.

Bueno, O. (1999) “Empiricism, Conservativeness and Quasi-truth”, Philosophy

of Science, 66, S474-85.

342

Bueno, O. (2001) “Review Symposium: The Quests of a Realist”, Metascience,

10(3), 360-6.

Bull, H. ([1977] 2002) The Anarchical Society: A Study of Order in World

Politics, 3rd edition. New York: Palgrave.

Bunge, M. (1996) Finding Philosophy in Social Science. London: Yale

University Press.

Buzan, B. (1995) “The Level of Analysis Problem in International Relations

Reconsidered”, in K. Booth & S. Smith (eds.) International Relations

Theory Today. Cambridge: Polity Press.

Buzan, B. (2006) “Series Editor’s Preface”, in S. Guzzini & A. Leander (eds.)

Constructivism and International Relations – Alexander Wendt and his

critics. London, New York: Routledge.

Callender, C. & Huggett, N. (eds.) (2001) Physics Meets Philosophy at the

Planck Scale: Contemporary Theories in Quantum Gravity. Cambridge:

Cambridge University Press.

Callender, C. & Magnus, P.D. (2004) “Realist Ennui and the Base Rate

Fallacy”, Philosophy of Science, 71(3), 320-38.

Cao, T.Y. (2003a) "Structural realism and the interpretation of quantum field

theory", Synthese, 136(1), 3-24.

Cao, T.Y. (2003b) "Can we dissolve physical entities into mathematical

structures?", Synthese, 136(1), 55-71.

Cao, T.Y. (2010) From Current Algebra to Quantum Chromodynamics - A

Case for Structural Realism. Cambridge: Cambridge University Press.

Carlsnaes, W. (1992) “The Agency-Structure Problem in Foreign Policy

Analysis”, International Studies Quarterly, 36(3), 245-70.

343

Carlsnaes, W. (1994) “In Lieu of a Conclusion: Compatibility and the Agency-

Structure Issue in Foreign Policy Analysis” in W. Carlsnaes & S. Smith

(eds.) European Foreign Policy – The EC and Changing Perspectives in

Europe. London: Sage Publications.

Carlsnaes, W. (2002) “Foreign Policy” in W. Carlsnaes, T. Risse & B.A.

Simmons (eds.) Handbook of International Relations. London: Sage

Publications.

Carlsnaes, W. (2006) “European Foreign Policy” in K.E. Jørgensen, M.A.

Pollack & B. Rosamond (eds.) Handbook of European Politics Union.

London: Sage Publications.

Carlsnaes, W. & Smith, S. (eds.) (1994) European Foreign Policy – The EC

and Changing Perspectives in Europe. London: Sage Publications.

Carlsnaes, W., Risse, T. & Simmons, B.A. (eds.) (2002) Handbook of

International Relations. London: Sage Publications.

Carman, C.C. (2005) “Discussion: The Electrons of the Dinosaurs and the

Center of the Earth: Comments on D. D. Turner’s ‘The Past vs. The Tiny:

Historical Science and the Abductive Arguments for Realism’”, Studies in

the History and Philosophy of Science, 36, 171-3.

Carr, E.H. ([1961] 2001) The Twenty Years’ Crisis – With a new introduction

by Michael Cox. London, New York: Palgrave.

Carrier, M. (1991) “What is Wrong with the Miracle Argument?”, Studies in the

History and Philosophy of Science, 22(1), 23-36.

Carrier, M. (1993) “What is Right with the Miracle Argument: Establishing a

Taxonomy of Natural Kinds”, Studies in the History and Philosophy of

Science, 24(3), 391-409.

344

Cartwright, N. (1983) How the Laws of Physics Lie. Oxford: Clarendon Press.

Castoriadis, C. (2011) Postscript on Insignificance – Dialogues with Cornelius

Castoriadis. London, New York: Continuum International Publishing

Group.

Chakravartty, A. (2001) “The Semantic or Model-Theoretic View of Theories

and Scientific Realism”, Synthese, 127(3), 325-45.

Chakravartty, A. (2004) “Structuralism as a Form of Scientific Realism”,

International Studies in the Philosophy of Science, 18 (2-3), 151-71.

Chalmers, D.J. (1995) “Minds, Machines and Mathematics” – Book Review [of

Penrose 1994a], PSYCHE – An interdisciplinary journal of research on

consciousness, available online at http://www.theassc.org/files/assc/2331.pdf

Chang, H. (2001) “How to take realism beyond foot-stamping”, Philosophy,

76(1), 5-30.

Chang, H. (2003) “Preservative Realism and its Discontents: Revisiting

Caloric”, Philosophy of Science, 70(5), 902-12.

Chernoff, F. (2002) “Scientific Realism as a Meta-Theory of International

Politics”, International Studies Quarterly, 46(2), 189-207.

Chernoff, F. (2005) The Power of International Theory: Reforging the Link to

Foreign Policy-making Through Scientific Inquiry. London: Routledge.

Chernoff, F. (2007) “Critical Realism, Scientific Realism, and International

Relations Theory”, Millennium: Journal of International Studies, 35(2), 399-

416.

Chernoff, F. (2009a) “Defending foundations for International Theory”,

International Theory, 1(3), 466-77.

345

Chernoff, F. (2009b) “The ontological fallacy: a rejoinder on the status of

scientific realism in international relations”, Review of International Studies,

35(2), 371-95.

Cheyne, C. & Worrall J. (eds.) (2006) Rationality and Reality: Conversations

with Alan Musgrave. Dordrecht: Springer.

Churchland, P.M. & Hooker, C.A. (eds.) ([1985] 2000) Images of Science: Essays

on Realism and Empiricism, with a Reply from Bas C. van Fraassen. Chicago:

University of Chicago Press.

Clarke, S. (2001) “Defensible Territory for the Entity Realism”, British Journal

for the Philosophy of Science, 52(4), 701-22.

Clarke, S. & Lyons D.T. (eds.) (2002) Recent Themes in the Philosophy of

Science: Scientific Realism and Common Sense. Dordrecht: Kluwer.

Cohen, R.S., Hilpinen, R. & Renzong, Q. (eds.) (1996) Realism and Anti-Realism

in the Philosophy of Science. Dordrecht: Kluwer.

Collier, A. (1994) Critical Realism: An Introduction to Roy Bhaskar's

Philosophy, London. New York: Verso.

Collingwood, R.G. ([1946]1973) The Idea of History. Oxford: Oxford

University Press.

Collins, H.M. ([1985] 1992) Changing Order: Replication and Induction in

Scientific Practice. Chicago: University of Chicago Press.

Colodny, R.G. (ed.) (1970) The Nature and Function of Scientific Theories.

Pittsburgh, PA: University of Pittsburgh Press.

Colyvan, M. (2001) The Indispensability of Mathematics. Oxford: Oxford

University Press.

346

Copeland, J. (1998) “Turing’s O-machines, Searle, Penrose and the Brain”,

Analysis, 58(2), 128-138.

Craig, W. (1956) “Replacement of Auxiliary Expressions”, Philosophical

Review, 65, 38-50.

Creath, R. (1985) “Taking Theories Seriously”, Synthese, 62, 317-45.

Crick, F. (1995) The Astonishing Hypothesis, the Scientific Search for the Soul.

New York: Simon and Schuster.

Cummiskey, D. (1992) “Reference Failure and Scientific Realism: a Response

to the Meta-Induction”, British Journal for the Philosophy of Science, 43(1),

21-40.

Curd, M. & Cover, J.A. (eds.) (1998) The Philosophy of Science: The Central

Issues. London, New York: W.W. Norton & Co.

Curtis, S. & Koivisto, M. (2010) “Towards a ‘Second Debate’? Rethinking the

Relationship between Science and History in International Theory”,

International Relations, 24(4), 433-455.

Da Costa, N.C.A. & French, S. (2003) Science and Partial Truth. Oxford:

Oxford University Press.

Davidson, D. (1963) “Actions, Reasons, and Causes”, Journal of Philosophy,

60(23), 685-700 • Also in A.R. White (ed.) The Philosophy of Action.

Oxford: Oxford University Press.

Davidson, D. & Harman G. (eds.) (1972) The Semantics of Natural Language.

Dordrecht: D. Reidel.

Dawid, R. (2007) "Scientific Realism in the Age of String Theory", Physics and

Philosophy, ISSN: 1863-7388, available online at https://eldorado.tu-

dortmund.de/bitstream/2003/24724/1/011.pdf .

347

De Regt, H. W. (2005) “Scientific Realism in Action: Molecular Models and

Boltzmann’s Bildtheorie”, Erkenntnis, 63(2), 205-30.

Dessler, D. (1989) “What’s at stake in the agent-structure debate”, International

Organization, 43(3), 441-73.

Detlefsen, M. (1998) “Mind in the Shadows” (Essay Review), Studies in the

History and Philosophy of Modern Physics, 29(1), 123-136.

Devitt, M. (1981) Designation. New York: Columbia University Press.

Devitt, M. ([1984] 1991) Realism and Truth, 2nd revised edition. Oxford:

Blackwell.

Devitt, M. (1999) “A Naturalistic Defence of Realism”, in S.D. Hales (ed.)

Metaphysics - Contemporary Readings. Belmont, CA: Wadsworth

Publishing Company.

Devitt, M. (2002) “Underdetermination and Realism”, Philosophical Issues, 12,

26-50.

Devitt, M. (2005) “Scientific Realism”, in F. Jackson, & S. Michael (eds.) The

Oxford Handbook of Contemporary Philosophy. Oxford: Oxford University

Press.

Dewdney, A. K. (1989) “A Pandora’s box of minds, machines and

metaphysics”, Computer Recreations Column, Scientific American,

December, 90-93.

Dicken, P. (2006) “Can the Constructive Empiricist be a Nominalist? Quasi-

truth, Commitment and Consistency”, Studies in the History and Philosophy

of Science, 37(2), 191-209.

Dieks, D. (ed.) (2006) The Ontology of Spacetime, vol. 1. Amsterdam: Elsevier.

Dieks, D. (ed.) (2008) The Ontology of Spacetime, vol. 2. Amsterdam: Elsevier.

348

Dilthey, William (1988), Introduction to the Human Sciences: An Attempt to

Lay a Foundation for the Study of Society and History, Betanzos, R. J. (tr.,

int.). London: Harvester Wheatsheaf.

Doty, R. L. (2000) “Desire all the way down”, Review of International Studies,

26(1), 137-39.

Douven, I. (2001) “Review Symposium: The Quests of a Realist”, Metascience,

10(3), 354-59.

Douven, I. (2005) “Evidence, Explanation, and the Empirical Status of

Scientific Realism”, Erkenntnis, 63, 253-91.

Douven, I. & van Brakel, J. (1995) “Is Scientific Realism an Empirical

Hypothesis?”, Dialectica, 49(1), 4-14.

Dray, W.H. (1964) Philosophy of History. New Jersey: Prentice-Hall, Inc.

Duhem, P. ([1906] 1954) The Aim and Structure of Physical Theory. Princeton,

NJ: Princeton University Press.

Duhem, P. ([1908] 1969) To Save the Phenomena. Chicago: University of

Chicago Press.

Dunne, T., Kurki, M. & Smith, S. (eds.) (2007) International Relations Theory

– Discipline and Diversity, 2nd edition. Oxford: Oxford University Press.

Earman, J. (1986) A Primer on Determinism. Dordrecht: Reidel.

Earman, J. (ed.) (1992) Inference, Explanation, and Other Frustrations.

Berkeley, Los Angeles: University of California Press.

Earman, J. (1993) “Underdetermination, Realism and Reason”, Midwest Studies

in Philosophy, 18(1), 19-38.

Economou, S.E. (2012), “Putting a spin on photon entanglement”, Nature, 491,

343-344.

349

Einstein, A., Podolsky B., & Rosen, N. (1935) "Can Quantum-Mechanical

Description of Physical Reality Be Considered Complete?", Physical

Review, 47, 777-780.

Eun, Y-S. (2012) “Why and how should we go for a multi-causal analysis in

the study of foreign policy? (Meta-)theoretical rationales and methodological

rules”, Review of International Studies, 38(4), 763-783.

Evans, R.J. (2000) In Defence of History. London: Granta Books.

Faris, W. (1996) Book Review [of Penrose 1994a], Notices of the American

Mathematical Society, 43(2), 203-208.

Feferman, S. (1995) “Penrose’s Gödelian Argument” – Book Review [of

Penrose 1994a], PSYCHE – An interdisciplinary journal of research on

consciousness), available online at

http://stuff.mit.edu/afs/athena/dept/libdata/applications/ejournals/b/n-

z/Psyche/2/psyche-95-2-7-shadows-5-feferman, and at

http://math.stanford.edu/~feferman/papers/penrose.pdf.

Feigl, H., Scriven, M. & Maxwell, G. (eds.) (1958) Minnesota Studies in the

Philosophy of Science, vol.2: Concepts, Theories and the Mind-Body

Problem. Minneapolis: University of Minnesota Press.

Feigl, H. & Maxwell, G. (eds.) (1962) Minnesota Studies in the Philosophy of

Science, vol.3: Scientific Explanation, Space, and Time. Minneapolis:

University of Minnesota Press.

Feynman R.P. (2007) Six Easy Pieces: Fundamentals of Physics Explained.

New York: Penguin.

Feynman, R. P., Leighton, R.B., and Sands, M. ([1965] 1989) The Feynman

Lectures on Physics, Volume III. Massachusetts: Addison – Wesley.

350

Fine, A. (1986a) The Shaky Game. Chicago: The University of Chicago Press.

Fine, A. (1986b) “Unnatural Attitudes: Realist and Instrumental Attachments to

Science”, Mind, 95(378), 149-79.

Fine, A. (1991) “Piecemeal Realism”, Philosophical Studies, 61(1-2), 79-96.

Fine, A. (2001) “The Scientific Image Twenty Years Later”, Philosophical

Studies, 106(1-2), 107-22.

Floyd, R. (2007) Book Review [of Wight 2006], International Affairs, 83(1),

188.

Fogelin, R.J. (1993) "Hume's scepticism", in D.F. Norton (ed.) The Cambridge

Companion to Hume. Cambridge: Cambridge University Press.

Franklin, A. (1986) The Neglect of Experiment. Cambridge: Cambridge

University Press.

Franzén, T. (2005) Gödel’s Theorem – An Incomplete Guide to its Use and

Abuse. Wellesley: A K Peters.

French, S. and Ladyman, J. (2003) "Remodelling structural realism: Quantum

physics and the metaphysics of structure", Synthese, 136(1), 31-56.

Fuller, S. (1989) Philosophy of Science and its Discontents. Boulder, Colorado:

Westview Press.

Fuller, S. (1991a) “Naturalized Epistemology Sublimated: Rapprochement

without the Ruts”, Studies in the History and Philosophy of Science, 22(2),

277-93.

Fuller, S. (1991b) Social Epistemology. Bloomington, IN: University of Indiana

Press.

Gadamer, H.G. (1975) Truth and Method. London: Sheed & Ward.

351

Gal, O. (2002) “Constructivism for Philosophers (Be it a Remark on Realism)”,

Perspectives on Science, 10(4), 523-49.

Galison, P. (1987) How Experiments End. Chicago: University of Chicago

Press.

Galison, P. (1988) “History, Philosophy, and the Central Metaphor”, Science in

Context, 2(1), 197-212.

Gasper, P. & Trout, J.D. (eds.) (1991) The Philosophy of Science. Cambridge,

Massachusetts: MIT Press.

Gelfert, A. (2003) “Manipulative Success and the Unreal”, International

Studies in the Philosophy of Science, 17(3), 245-63.

Giddens, A. (1979) Central Problems in Social Theory. London: Hutchinson.

Giddens, A. (1984) The Constitution of Society: Outline of the Theory of

Structuration. Cambridge: Polity Press.

Giere, R. (1998) Science Without Laws. Chicago: The University of Chicago

Press.

Giere, R. (2005) “Scientific Realism: Old and New Problems”, Erkenntnis,

63(2), 149-65.

Glock, H-J. (1996) A Wittgenstein Dictionary. Oxford: Blackwell.

Glymour, C. (1980) Theory and Evidence. Princeton, NJ: Princeton University

Press.

Glymour, C. ([1992]1999) “Realism and the Nature of Theories”, in M.H.

Salmon et al. Introduction to the Philosophy of Science. Indianapolis,

Cambridge: Hackett Publishing Company.

Gofas, A. (2007) The Ideas Debate in International and European Studies: A

Critical Re-Appraisal, University of Warwick, PhD Thesis (unpublished).

352

Gooding, D. (1990) Experiment and the Making of Meaning. Dordrecht:

Kluwer.

Gooding, D., Pinch, T. & Schaffer, S. (eds.) (1989) The Uses of Experiment.

Cambridge: Cambridge University Press.

Gould, H. D. (1998) “What is at stake in the Agent-Structure Debate?”, in V.

Kubálkova, N. Onuf, N. & P. Kowert (eds.) International Relations in a

Constructed World. London, New York: M. E. Sharpe.

Griffiths, M., Roach, S.C., and Solomon, M.S. ([1999] 2009) Fifty Key

Thinkers in International Relations, 2nd edition. London, New York:

Routledge.

Griffiths, R. (2002) Consistent Quantum Theory. Cambridge: Cambridge

University Press.

Groff, R. (2000) “The Truth of the Matter: Roy Bhaskar’s Critical Realism and

the Concept of Alethic Truth”, Philosophy of the Social Sciences, 30(3),

407-35.

Groom, A.J.R. & Light, M. (eds.) (1994) Contemporary International

Relations: A Guide to Theory. London: Pinter Publishers Ltd.

Gunnell, J. G. (1995) “Realizing theory: the Philosophy of Science revisited”,

Journal of Politics, 57(4), 923-40.

Gutting, G. (1982) “Scientific Realism versus Constructive Empiricism: a

Dialogue”, The Monist, 65, 336-49 • Also in P.M. Churchland & C.A.

Hooker (eds.) Images of Science: Essays on Realism and Empiricism, with a

Reply from Bas C. van Fraassen. Chicago: University of Chicago Press.

Guzzini, S. & Leander, A. (eds.) (2006a) Constructivism and International

Relations – Alexander Wendt and his critics. London, New York: Routledge.

353

Guzzini, S. & Leander, A. (2006b) “Preface”, in S. Guzzini & A. Leander (eds.)

Constructivism and International Relations – Alexander Wendt and his

critics. London, New York: Routledge.

Guzzini, S. & Leander, A. (2006c) “Wendt’s constructivism – A relentless

quest for synthesis”, in S. Guzzini & A. Leander (eds.) Constructivism and

International Relations – Alexander Wendt and his critics. London, New

York: Routledge.

Haack, S. (1987) “Scientific Realism”, Synthese, 73(2), 275-300.

Habermas, J. (1973) Theory and Practice. Boston: Beacon.

Habermas, J. (1978) Knowledge and Human Interests, 2nd edition. London:

Heinemann.

Habermas, J. (1984) The Theory of Communicative Action, Vol. I: Reason and

the Rationalization of Society. Boston: Beacon.

Habermas, J. (1987) The Theory of Communicative Action, Vol. II: Lifeworld

and System. Boston: Beacon.

Hacker, P.M.S. (1996), Wittgenstein: Mind and Will (Vol.4 of An Analytical

Commentary on the Philosophical Investigations). Oxford: Blackwell.

Hacking, I. (1983) Representing and Intervening: Introductory Topics in the

Philosophy of Natural Science. Cambridge: Cambridge University Press.

Hales, S.D. (ed.) (1999) Metaphysics - Contemporary Readings. Belmont, CA:

Wadsworth Publishing Company.

Halliday, F. (1994) Rethinking International Relations. Hampshire, London:

Macmillan.

354

Halliday, F. (1996) “The future of international relations: fears and hopes”, in S.

Smith, K. Booth & M. Zalewski (eds.) International Theory: Positivism and

Beyond. Cambridge: Cambridge University Press.

Halliday, F. (2000) “International relations theory and the post-cold war

period”, METU Studies in Development, 27(3-4), 235-63.

Hameroff, S. (1994) “Quantum Coherence in Microtubules: A Neural Basis for

Emergent Consciousness?”, Journal of Consciousness Studies, 1(1), 91-118.

Hameroff, S. (2001) “Biological Feasibility of Quantum Approaches to

Consciousness”, in P. Van Loocke (ed.) The Physical Basis of

Consciousness. Philadelphia: John Benjamins.

Hameroff, S. and Penrose, R. (1996) “Conscious Events as Orchestrated

Space-Time Selections”, Journal of Consciousness Studies, 3(1), 36-53.

Hanson, N. R. (1958) Patterns of Discovery. Cambridge: Cambridge University

Press.

Hardin, C. & Rosenberg, A. (1982) “In Defence of Convergent Realism”,

Philosophy of Science, 49(4), 604-15.

Harré, R. (1996) “From Observability to Manipulability: Extending the

Inductive Arguments for Realism”, Synthese, 108(2), 137-55.

Hay, C. (2002) Political Analysis – A Critical Analysis. Hampshire, New York:

Palgrave Macmillan.

Hay, C. (2005) “Making Hay…or Clutching at Ontological Straws? Notes on

Realism, ‘As-if-Realism’ and Actualism”, Politics, 25(1), 39-45.

Healey, R. (2009) What’s Real – The Conceptual Foundations of

Contemporary Gauge Theories. Oxford: Oxford University Press.

355

Heidegger, M. (1997) Kant and the Problem of Metaphysics. Indiana: Indiana

University Press.

Hempel, C. (1958) “The Theoretician’s Dilemma: A Study in the Logic of

Theory Construction”, in H. Feigl, M. Scriven & G. Maxwell (eds.)

Minnesota Studies in the Philosophy of Science, vol.2: Concepts, Theories

and the Mind-Body Problem. Minneapolis: University of Minnesota Press.

Hempel, C. (1965) Aspects of Scientific Explanation and Other Essays in the

Philosophy of Science. New York: The Free Press.

Herborth, B. (2012) “Theorising theorising: Critical Realism and the quest for

certainty”, Review of International Studies, 38(1), 235-51.

Hill, C. (ed.) (1983) National Foreign Policies and European Political

Cooperation. London: George Allen & Unwin.

Hill, C. (1989) “1939: the origins of liberal realism”, Review of International

Studies, 15(4), 319-328.

Hill, C. (1991) Cabinet decisions on foreign policy – The British experience

October 1938-June 1941, LSE Monographs in International Studies.

Cambridge: Cambridge University Press.

Hill, C. (ed.) (1996a) The actors in Europe’s Foreign Policy. London:

Routledge.

Hill, C. (1996b) “World opinion and the empire of circumstance”,

International Affairs, 72(1), 109-31.

Hill, C. (1999) “‘Where are we going?’ – International Relations and the voice

from below”, Review of International Studies, 25 (1), 107-22.

Hill, C. (2003) The Changing Politics of Foreign Policy. Hampshire, New

York: Palgrave Macmillan.

356

Hill, C. (2007) “Bringing War Home: Foreign Policy-Making in Multicultural

Societies”, International Relations, 21(3), 259-83.

Hill, C. (2011) “Changing the world? - The problem of action in international

politics”, Cambridge Review of International Affairs, 24(1), 81-98.

Hill, C. & Smith, M. (eds.) (2011a), International Relations and the European

Union, 2nd edition. Oxford: Oxford University Press.

Hill, C. & Smith, M. (2011b) “International Relations and the European Union:

Themes and Issues”, in C. Hill & M. Smith (eds.) International Relations

and the European Union, 2nd edition. Oxford: Oxford University Press.

Hitchcock, C. (ed.) (2004) Contemporary Debates in the Philosophy of Science.

Oxford: Blackwell Publishing.

Hoefer, C. & Rosenberg, A. (1994) “Empirical Equivalence,

Underdetermination and Systems of the World”, Philosophy of Science,

61(4), 592-607.

Hollis, M. (1992) “Social Thought and Social Action”, in E. McMullin (ed.)

The Social Dimensions of Science. Notre Dame: University of Notre Dame

Press.

Hollis, M. (1994) The philosophy of social science – An introduction.

Cambridge: Cambridge University Press.

Hollis, M. & Smith, S. (1990) Explaining and Understanding International

Relations. Oxford: Clarendon.

Hollis, M. & Smith, S. (1991) “Beware of gurus: structure and action in

international relations”, Review of International Studies, 17(4), 393-410.

Hollis, M. & Smith, S. (1992) “Structure and action: further comment”, Review

of International Studies, 18(2), 187-88.

357

Hollis, M. & Smith, S. (1994) “Two stories about structure and agency”,

Review of international Studies, 20(3), 241-51.

Hollis, M. & Smith, S. (1996) “A response: why epistemology matters in

international theory”, Review of International Studies, 22(1), 111-16.

Holsti, K. J. (1991) Change in the International System – Essays on the Theory

and Practice of International Relations. Aldershot, Hunts: Edward Elgar.

Horwich, P. (1998) Truth, 2nd edition. Oxford: Clarendon Press.

Howson, C. (2000) Hume’s Problem: Induction and the Justification of Belief.

Oxford: Clarendon Press.

Hoyningen-Huene, P. (1993) Reconstructing Scientific Revolutions: Thomas S.

Kuhn’s Philosophy of Science. Chicago: University of Chicago Press.

Hull, D., Forbes, M. & Burian, R.M. (eds.) (1995) PSA 1994, vol. 1. East

Lansing, MI: Philosophy of Science Association.

Hutchings, K. (1999) International Political Theory. London: Polity.

Irwin, T. (1988) Aristotle’s First Principles. Oxford: Oxford University Press.

Irwin, T. (1995) Plato's Ethics. New York: Oxford University Press.

Jabri, V. & Chan, S. (1996) “The ontologist always rings twice: two more

stories about structure and agency in reply to Hollis and Smith”, Review of

International Studies, 22(1), 107-10.

Jackson, F. & Michael S. (eds.) (2005) The Oxford Handbook of Contemporary

Philosophy. Oxford: Oxford University Press.

Jackson, P.T. (2001), Book Review Essay [of Wendt 1999]: “Constructing

Thinking Space — Alexander Wendt and the Virtues of Engagement”,

Cooperation and Conflict, 36(1), 109-20.

358

Jackson, P.T. (2004a) “Forum Introduction – Is the state a person? Why should

we care?”, Review of International Studies, 30(2), 255-58.

Jackson, P.T. (2004b) “Hegel’s House, or ‘People are states too’”, Review of

International Studies, 30(2), 281-87.

Jackson, P.T. (2008a) “Ontological Investigations” – Book Review [of Wight

2006], Cooperation and Conflict, 43(3), 341-60.

Jackson, P.T. (2008b) “Foregrounding ontology: dualism, monism and IR

theory”, Review of International Studies, 34(1), 129-53.

Jackson, P.T. (2009) “A faulty solution to a False(ly characterized) problem: a

comment on Monteiro and Ruby”, International Theory, 1(3), 455-65.

Jackson, P.T. (2011) The Conduct of Inquiry in International Relations. New

York: Routledge.

Jammer, M. (1974) The Philosophy of Quantum Mechanics: The

Interpretations of Quantum Mechanics in Historical Perspective. New

York: John Wiley & Sons.

Jessop, B. (1990) State Theory: Putting Capitalist States in their Place.

Cambridge: Polity.

Jessop, B. (2001) “Capitalism, the Regulation Approach, and Critical Realism”,

in A. Brown, S. Fleetwood & J.M. Roberts (eds.) Critical Realism and

Marxism. London: Routledge.

Jessop, B. (2002) “Critical Realism and the Strategic-Relational Approach”,

New Formations, 56, 40-53.

Jessop, B. (2007) State Power. Cambridge: Polity.

Jessop, B. (2010) “World Market, World State, world Society: Marxian

Insights and Scientific Realist Interrogations”, in J. Joseph & C. Wight

359

(eds.) Scientific Realism and International Relations. New York: Palgrave

Macmillan.

Johnson, G. (1990) Book Review [of Penrose 1989], The Sciences, New York

Academy of Sciences, July- August, 45- 49.

Jørgensen, K.E., Pollack, M.A. & Rosamond, B. (eds.) (2006) Handbook of

European Politics Union. London: Sage Publications.

Joseph, J. (2002) Hegemony: A Realist Analysis. London, New York:

Routledge.

Joseph, J. (2007) “Philosophy in International Relations: A Scientific Realist

Approach”, Millennium: Journal of International Studies, 35(2), 345-59.

Joseph, J. (2008) “Hegemony and the Structure-Agency Problem in

International Relations: a Scientific Realist Contribution”, Review of

International Studies, 34(1), 109-28.

Joseph, J. (2010) “The International as Emergent: Challenging Old and New

Orthodoxies in International Relations Theory”, in J. Joseph & C. Wight

(eds.) Scientific Realism and International Relations. New York: Palgrave

Macmillan.

Joseph, J. & Wight, C. (eds.) (2010) Scientific Realism and International

Relations. New York: Palgrave Macmillan.

Josephson, B. (1994) “To Err is Mechanical?” – Book Review [of Penrose

1994a & the Journal of Consciousness Studies], The Times Higher

Education, November the 4th, 23-24.

Kant, I. (1997) Prolegomena to Any Future Metaphysics (edited by G.

Hatfield). Cambridge: Cambridge University Press.

360

Kant, I. (1999) Critique of Pure Reason. Cambridge: Cambridge University

Press.

Käpylä, J. & Mikkola, H. (2010) “A Critical Look at Critical Realism: Some

Observations on the Problems of Meta-theory”, World Political Science

Review, 6(1), 1-37.

Keeley, J.F. (2007) “To the Pacific? Alexander Wendt as Explorer”,

Millennium: Journal of International Studies, 35(2), 417-30.

Kessler, O. (2007) “From Agents and Structures to Minds and Bodies: of

Supervenience, Quantum and the Linguistic Turn”, Journal of International

Relations and Development, 10(3), 243-271.

Kessler, O. (2012) “On logic, intersubjectivity, and meaning: is reality an

assumption we just don’t need”, Review of International Studies, 38(1), 253-

65.

Keukeleire, S. & MacNaughtan, J. (2008) The Foreign Policy of European

Union. New York: Palgrave Macmillan.

Kim, J. (1998) Philosophy of Mind. Boulder: Westview Press.

King, G., Keohane, R.O. & Verba, S. (1994) Designing Social Inquiry.

Princeton, New Jersey: Princeton University Press.

Kitcher, P. (1993) The Advancement of Science: Science without Legend,

Objectivity without Illusions. Oxford: Oxford University Press.

Kitcher, P. (2001a) “Real Realism: The Galilean Strategy”, Philosophical

Review, 110, 151-97.

Kitcher, P. (2001b) Science, Truth, and Democracy. New York: Oxford

University Press.

361

Kivinen, O. & Piironen, T. (2004) “The relevance of ontological commitments

in social science: realist and pragmatist viewpoints”, Journal of the Theory

of Social Behavior, 34(3), 231-48.

Klein, S.A. (1995) “Is quantum mechanics relevant to understanding

consciousness? – Book Review [of Penrose 1994a], PSYCHE – An

Interdisciplinary Journal of Research on Consciousness, available online at

http://www.theassc.org/files/assc/2326.pdf .

Klein, S.A. (1961) “Minds, machines and Gödel”, Philosophy, 36(137), 112-

27.

Knorr-Cetina, K. (1981) The Manufacture of Knowledge. Oxford & New York:

Pergamon Press.

Knorr-Cetina, K. (1999) Epistemic Cultures: How the Sciences Make Knowledge.

Cambridge, Massachusetts: Harvard University Press.

Knudsen, O. F. (1994) “Context and Action in the Collapse of the Cold War

European System” in W. Carlsnaes & S. Smith, S. (eds.) European Foreign

Policy – The EC and Changing Perspectives in Europe. London: Sage

Publications.

Koivisto, M. (2010) “State Theory in International Relations: Why Realism

Matters”, in J. Joseph & C. Wight (eds.) Scientific Realism and International

Relations. New York: Palgrave Macmillan.

Koivisto, M. (2012) Normative State Power in International Relations. Oxford:

Oxford University Press.

Kořan, M. (2007) Book Review [of Wight 2006], Journal of International

Relations and Development, 10, 324-26.

362

Krasner, S. D. (2000) “Wars, Hotel Fires, and Plane Crashes”, Review of

International Studies, 26(1), 131-6.

Kratochwil, F. (1989) Rules, Norms and decisions. Cambridge: Cambridge

University Press.

Kratochwil, F. (2000) “Constructing a new orthodoxy? – Wendt’s Social

Theory of International Politics and the constructivist challenge”,

Millennium: Journal of International Studies, 29(1), 73-101 • Also in S.

Guzzini & A. Leander (eds.) Constructivism and International Relations –

Alexander Wendt and his critics. London, New York: Routledge.

Kratochwil, F. (2006) “History, Action and Identity: Revisiting the ‘Second’

Great Debate and Assessing its Importance for Social Theory”, European

Journal of International Relations, 12(1), 5-29.

Kratochwil, F. (2007a) “Of false promises and good bets: a plea for a pragmatic

approach to theory building (the Tartu lecture)”, Journal of International

Relations and Development, 10(1), 1-15.

Kratochwil, F. (2007b) “Of communities, gangs, historicity and the problem of

Santa Claus: replies to my critics”, Journal of International Relations and

Development, 10(1), 57-78.

Kraut, R. (1991) Aristotle on the Human Good. Princeton: Princeton University

Press.

Kripke, S. (1972) “Naming and Necessity”, in D. Davidson & G. Harman (eds.)

(1972) The Semantics of Natural Language. Dordrecht: D. Reidel.

Kubálkova, V., Onuf, N. & Kowert, P. (eds.) (1998) International Relations in

a Constructed World. London, New York: M.E. Sharpe.

363

Kuhlmann, M. et al. (eds.) Ontological Aspects of Quantum Field Theory.

Singapore: World Scientific.

Kuhn, T.S. ([1962]1970) The Structure of Scientific Revolutions, 2nd enlarged

edition. Chicago: University of Chicago Press.

Kukla, A. (1993) “Laudan, Leplin, Empirical Equivalence and

Underdetermination”, Analysis, 53(1), 1-7.

Kukla, A. (1994a) “Non-Empirical Theoretical Virtues and the Argument from

Underdetermination”, Erkenntnis, 41(2), 157-70.

Kukla, A. (1994b) “Scientific Realism, Scientific Practice and the Natural

Ontological Attitude”, British Journal for the Philosophy of Science, 45(4),

955-75.

Kukla, A. (1995) “The Two Antirealisms of Bas van Fraassen”, Studies in the

History and Philosophy of Science, 26(3), 431-54.

Kukla, A. (1996a) “Antirealist Explanations of the Success of Science”,

Philosophy of Science, 63 (Supplement), S298 - S305.

Kukla, A. (1996b) “Does Every Theory Have Empirically Equivalent Rivals?”,

Erkenntnis, 44(2), 137-66.

Kukla, A. (1998) Studies in Scientific Realism. Oxford: Oxford University

Press.

Kukla, A. (2000) “Theoreticity, Underdetermination and the Disregard for

Bizarre Scientific Hypotheses”, Philosophy of Science, 68(1), 21-35.

Kurki, M. (2006) “Causes of a divided discipline: rethinking the concept of

cause in IR”, Review of International Studies, 32(2), 189-216.

Kurki, M. (2007) “Critical Realism and Causal Analysis in International

Relations”, Millennium: Journal of International Studies, 35(2), 361-78.

364

Kurki, M. (2008) Causation in International Relations-Reclaiming Causal

Analysis. Cambridge: Cambridge University Press.

Kurki, M. (2009) “The politics of the philosophy of science”, International

Theory, 1(3), 440-54.

Kurki, M. (2011) “The Limitations of Critical Edge: Reflections on Critical and

Philosophical IR Scholarship”, Millennium: Journal of International Studies,

40(1), 129-146.

Kurki, M. & Wight, C. (2007) “International Relations and Social Science”, in

T. Dunne, M. Kurki, & S. Smith (eds.) International Relations Theory –

Discipline and Diversity, 2nd edition. Oxford: Oxford University Press.

Kurki, M. & Sinclair, A. (2010) “Hidden in plain sight: Constructivist treatment

of social context and its limitations”, International Politics, 47(1), 1-25.

Ladyman, J. (1998) “What is Structural Realism?”, Studies in the History and

Philosophy of Science, 29(3), 409-24.

Ladyman, J. (2002) Understanding Philosophy of Science. London: Routledge.

Landau, L.D. and Lifshitz, E.M. (1977) Quantum Mechanics, 3rd edition.

Oxford: Pergamon Press.

Lange, M. (2002) “Baseball, Pessimistic Induction, and the Turnover Fallacy”,

Analysis, 62, 281-85.

Latham, A. (2012) “A Pluralistic Science of IR” – Book Review [of Jackson

2011], The Review of Politics, 74(2), 362-64.

Laudan, L. (1981) “A Confutation of Convergent Realism”, Philosophy of

Science, 48(1), 19-49 • Also in J. Leplin (ed.) Scientific Realism. Berkeley:

University of California Press.

365

Laudan, L. (1984) Science and Values. Berkeley: University of California

Press.

Laudan, L. (1990) “Demystifying Underdetermination”, in W. Savage (ed.)

Minnesota Studies in the Philosophy of Science, vol. 14: Scientific Theories.

Minneapolis: University of Minnesota Press • Also in M. Curd & J.A. Cover

(eds.) The Philosophy of Science: The Central Issues. London, New York:

W. W. Norton & Co.

Laudan, L. & Jarett L. (1991) “Empirical Equivalence and

Underdetermination”, Journal of Philosophy, 88, 448–72.

Lawson, G. (2007) Book Review [of Wight 2006], Millennium: Journal of

International Studies, 35(3), 773-75.

Lebow, R.N. (2003) The tragic vision of politics: ethics, interests and others.

Cambridge: Cambridge University Press.

Lebow, R.N. (2008) A cultural theory of International Relations. Cambridge:

Cambridge University Press.

Lebow, R.N. (2011) “Philosophy and International Relations” - Book Review

[of Joseph & Wight 2010 and Jackson 2011], International Affairs, 87(5),

1219-28.

Leeds, S. (1994) “Constructive Empiricism”, Synthese, 101(2), 187-221.

Leng, M. (2005) “Platonism and Anti-Platonism: Why Worry?”, International

Studies in the Philosophy of Science, 19(1), 65-84.

Leon, D. (2010) “Reductionism, Emergence and Explanation in International

Relations Theory”, in J. Joseph & C. Wight (eds.) Scientific Realism and

International Relations. New York: Palgrave Macmillan.

366

Leplin, J. (1997a) A Novel Defence of Scientific Realism. Oxford: Oxford

University Press.

Leplin, J. (1997b) “The Underdetermination of Total Theories”, Erkenntnis,

47, 203-15.

Leplin, J. (1979) “Reference and Scientific Realism”, Studies in the History and

Philosophy of Science, 10(4), 265-84.

Leplin, J. (ed.) (1984a) Scientific Realism. Berkeley: University of California

Press.

Leplin, J. (1984b) “Truth and Scientific Progress”, in J. Leplin (ed.) Scientific

Realism. Berkeley: University of California Press.

Leplin, J. (2000) “The Epistemological Status of Auxiliary Hypotheses: a Reply

to Douven”, Philosophical Quarterly, 50(202), 376-80.

Leplin, J. (2001) “Realism and Instrumentalism”, in W. Newton-Smith (ed.) A

Companion to the Philosophy of Science. Oxford: Blackwell Publishing.

Leplin, J. (2004) “A Theory’s Predictive Success Can Warrant Belief in the

Unobservable Entities it Postulates”, in C. Hitchcock (ed.) Contemporary

Debates in the Philosophy of Science. Oxford: Blackwell Publishing.

Levin, M. (1984) “What Kind of Explanation Is Truth?”, in J. Leplin (ed.)

Scientific Realism. Berkeley: University of California Press.

Lewis, P. (2001) “Why the Pessimistic Induction is a Fallacy”, Synthese, 129,

371-80.

Light, M. & Groom, A.J.R. (eds.) (1985) International Relations – A Handbook

of Current Theory. London: Pinter.

367

Lipton, P. (1993) “Is the Best Good Enough?”, Proceedings of the Aristotelian

Society, 93(2), 89-104 • Also in D. Papineau (ed.) The Philosophy of Science.

Oxford: Oxford University Press.

Lipton, P. (2001) “Review Symposium: The Quests of a Realist”, Metascience,

10(3), 347-53.

Lipton, P. (2005) “The Medawar Lecture 2004: The Truth About Science”,

Philosophical Transactions of the Royal Society B, 360 (1458), 1259-69.

Little, R. (2008) “History, Theory and Methodological Pluralism in the English

School”, in C. Navari (ed.) Theorising International Society: English School

Methods. New York: Palgrave Macmillan.

Lucas, J.R. (1970) The Freedom of Will. Oxford: Oxford University Press.

Luhmann, N. (1984) Soziale Systeme. Frankfurt am Main: Suhrkamp.

Luhmann, N. (1997) Die Gesselschaft der Gesselschaft. Frankfurt am Main:

Suhrkamp.

Machamer, P. & Silberstein M. (eds.) (2002) The Blackwell Guide to the

Philosophy of Science. Malden, MA: Blackwell.

Mackie, J.L. (1990) Ethics: Inventing Right and Wrong. London: Penguin.

Mäki, U. (2005) “Reglobalizing Realism by Going Local, or (How) Should our

Formulations of Scientific Realism be Informed about the Sciences?”,

Erkenntnis, 63, 231-51.

Manners, I. and Whitman, R. (eds.) (2000) The Foreign Policies of European

Union Member States. Manchester: Manchester University Press.

Maturana, H.R. & Varela, F.J. (1992) The Tree of Knowledge: The Biological

Roots of Human Understanding. London: Shambhala Publications.

368

Maudlin, T. (1995) “Between the motion and the act…” – Book Review [of

Penrose 1994a], PSYCHE – An Interdisciplinary Journal of Research on

Consciousness, available online at

http://stuff.mit.edu/afs/athena/dept/libdata/applications/ejournals/b/n-

z/Psyche/2/psyche-95-2-2-som-1-maudlin .

Maxwell, G. (1962) “The Ontological Status of Theoretical Entities”, in H.

Feigl, & G. Maxwell (eds.) Minnesota Studies in the Philosophy of Science,

vol.3: Scientific Explanation, Space, and Time. Minneapolis: University of

Minnesota Press • Also in M. Curd & J.A. Cover (eds.) The Philosophy of

Science: The Central Issues. London, New York: W. W. Norton & Co.

Maxwell, G. (1970) “Theories, Perception, and Structural Realism”, in R.G.

Colodny (ed.) The Nature and Function of Scientific Theories. Pittsburgh,

PA: University of Pittsburgh Press.

McAllister, J.W. (1993) “Scientific Realism and the Criteria for Theory-

choice”, Erkenntnis, 38, 203-22.

McAnulla, S. (2005) “Making Hay with Actualism? The Need for a Realist

Concept of Structure”, Politics, 25(1), 31-38.

McArthur, D. (2003) “Reconsidering Structural Realism”, Canadian Journal of

Philosophy, 33(4), 517-36.

McCarthy, J. (1995) “Awareness and Understanding in Computer Programs” –

Book Review [of Penrose 1994a], PSYCHE – An interdisciplinary journal

of research on consciousness, available online at

http://www.theassc.org/files/assc/2332.pdf .

McCullogh, D. (1995) “Can humans escape Gödel?” – Book Review [of

Penrose 1994a], PSYCHE – An Interdisciplinary Journal of Research on

369

Cconsciousness, available online at

http://www.theassc.org/files/assc/2327.pdf .

McDermott, D. (1995) “Penrose is Wrong” - Book Review [of Penrose 1994a],

PSYCHE – An Interdisciplinary Journal of Research on Consciousness,

available online at http://www.theassc.org/files/assc/2335.pdf .

McGowan, M.K. (1999) “The Metaphysics of Squaring Scientific Realism with

Referential Indeterminacy”, Erkenntnis, 50(1), 87-94.

McLeish, C. (2005) “Scientific Realism Bit by Bit: Part I. Kitcher on

Reference”, Studies in the History and Philosophy of Science, 36, 667-85.

McMichael, A. (1985) “Van Fraassen’s Instrumentalism”, British Journal for

the Philosophy of Science, 36(3), 257–72.

McMullin, E. (1984) “A Case for Scientific Realism”, in J. Leplin (ed.)

Scientific Realism. Berkeley: University of California Press • Also in Y.

Balashov & A. Rosenberg (eds.) Philosophy of Science – Contemporary

Readings. London: Routledge.

McMullin, E. (1987) “Explanatory Successes and the Theory of Truth”, in N.

Rescher (ed.) Scientific Inquiry in Philosophical Perspective. Lanham:

University Press of America.

McMullin, E. (1991) “Comment: Selective Anti-Realism”, Philosophical

Studies, 61(1-2), 97-108.

McMullin, E. (ed.) (1992) The Social Dimensions of Science. Notre Dame:

University of Notre Dame Press.

Mercado, R. (2009) “Keep muddling through?”, International Theory, 3(1),

478-87.

370

Micale, M.S. (2012) “Improvements” – Book Review [of Pinker 2011], The

Times Literature Supplement (TLS), 9th March.

Michel, T. (2009) “Pigs can’t fly, or can they? Ontology, scientific realism and

the metaphysics of presence in international relations”, Review of

International Studies, 35(2), 397-419.

Michel, T. (2012) “In Heidegger’s shadow: a phenomenological critique of

Critical Realism”, Review of International Studies, 38(1), 209-222.

Miller, R. (1987) Fact and Method. Princeton, NJ: Princeton University Press.

Monteiro, N.P. & Ruby, K.G. (2009a) “IR and the false promise of philosophical

foundations”, International Theory, 1(1), 15-48.

Monteiro, N.P. & Ruby, K.G. (2009b) “The promise of foundational prudence: a

response to our critics”, International Theory, 1(3), 499-512.

Mouzakitis, K. (2011) “Still Fashionable Yet Useless? Addressing Problems

with Research on the Europeanization of Foreign Policy”, Journal of

Common Market Studies, 49(3), 607-629.

Morgan, J. (2002) “Philosophical Realism in International Relations Theory –

Kratochwil’s Constructivist Challenge to Wendt”, Journal of Critical

Realism, 1(1), 95-118.

Morgan, M. (1999) “Models as Mediating Instruments”, in M. Morgan & M.

Morrison (eds.) Models as Mediators: Essays on the Philosophy of the Natural

and Social Sciences. Cambridge: Cambridge University Press.

Morgan, M. & Morrison, M. (eds.) (1999) Models as Mediators: Essays on the

Philosophy of the Natural and Social Sciences. Cambridge: Cambridge

University Press.

371

Morrison, M. (1990) “Theory, Intervention and Realism”, Synthese, 82(1), 1-

22.

Morrison, M. (1999) “Models as Autonomous Agents”, in M. Morgan & M.

Morrison (eds.) Models as Mediators: Essays on the Philosophy of the Natural

and Social Sciences. Cambridge: Cambridge University Press.

Mouzelis, N. (1991) Back to Sociological Theory: The Construction of Social

Orders. Hampshire, London: Macmillan.

Mouzelis, N. (1995) Sociological Theory: What went wrong?. London:

Routledge.

Mouzelis, N. (2008) Modern and Postmodern Social Theorizing: Bridging the

Divide. Cambridge: Cambridge University Press.

Murphy, N. (1990) “Scientific Realism and Postmodern Philosophy”, British

Journal for the Philosophy of Science, 41, 291-303.

Musgrave, A. (1985) “Realism versus Constructive Empiricism”, in P.M.

Churchland & C.A. Hooker (eds.) Images of Science: Essays on Realism and

Empiricism, with a Reply from Bas C. van Fraassen. Chicago: University of

Chicago Press • Also in M. Curd & J.A. Cover (eds.) The Philosophy of

Science: The Central Issues. London, New York: W.W. Norton & Co.

Musgrave, A. (1988) “The Ultimate Argument for Scientific Realism”, in R.

Nola (ed.) Relativism and Realism in Science. Dordrecht: Kluwer.

Musgrave, A. (1989) “NOA’s Ark —Fine for Realism”, Philosophical Quarterly,

39, 382-98 • Also in M. Curd & J.A. Cover (eds.) The Philosophy of Science:

The Central Issues. London, New York: W.W. Norton & Co.

372

Musgrave, A. (1996) “Realism, Truth and Objectivity”, in R.S. Cohen, R.

Hilpinen & Q. Renzong (eds.) Realism and Anti-Realism in the Philosophy of

Science. Dordrecht: Kluwer.

Nagel, T. (1972) “Aristotle on Eudaimonia”, Phronesis, 17(3), 252-259.

Navari, C. (ed.) (2008) Theorising International Society: English School

Methods. New York: Palgrave Macmillan.

Nelson, A. (1994) “How Could Scientific Facts be Socially Constructed?”,

Studies in the History and Philosophy of Science, 25(4), 535-47.

Neufeld, M.A. (1995) The Restructuring of International Relations Theory.

Cambridge: Cambridge University Press.

Neuhouser, F. (2003) Foundations of Hegel's Social Theory: Actualizing

Freedom. Massachusetts: Harvard University Press.

Neumann, I.B. (2004) “Beware of organicism: the narrative self of the state”,

Review of International Studies, 30(2), 259-67.

Neumann, I.B. & Wæver, O. (eds.) ([1997] 2001) The Future of International

Relations - Masters in the Making. London, New York: Routledge.

Newton-Smith, W. (1978) “The Underdetermination of Theory by Data”,

Proceedings of the Aristotelian Society (Supplementary Volume), 52, 71-91 •

Also in W. Newton-Smith (ed.) A Companion to the Philosophy of Science.

Oxford: Blackwell Publishing.

Newton-Smith, W. (ed.) (2001) A Companion to the Philosophy of Science.

Oxford: Blackwell Publishing.

Nicholson, M. (1985) “Methodology”, in M. Light & A.J.R. Groom (eds.)

International Relations – A Handbbok of Current Theory. London: Pinter.

373

Nicholson, M. (1996a) Causes and Consequences in International Relations: A

Conceptual Study. New York: Cassel Academic.

Nicholson, M. (1996b) “The continued significance of positivism?”, in S. Smith,

K. Booth, & M. Zalewski (eds.) International Theory: Positivism and Beyond.

Cambridge: Cambridge University Press.

Nicholson, M. & Bennett, P. (1994a) “The epistemology of international

relations”, in A.J.R. Groom & M. Light (eds.) Contemporary International

Relations: A Guide to Theory. London: Pinter Publishers Ltd.

Nicholson, M. & Bennett, P. (1994b) “Formal methods of analysis in IR”, in

A.J.R. Groom & M. Light (eds.) Contemporary International Relations: A

Guide to Theory. London: Pinter Publishers Ltd.

Niiniluoto, I. (1999) Critical Scientific Realism. Oxford: Oxford University Press.

Nola, R. (ed.) (1988) Relativism and Realism in Science. Dordrecht: Kluwer.

Nola, R. (1994) “There Are More Things in Heaven and Earth, Horatio, than are

Dreamt of in Your Philosophy: A Dialogue on Realism and Constructivism”,

Studies in the History and Philosophy of Science, 25, 689-27.

Nola, R. (2002) “Realism through Manipulation, and by Hypothesis”, in S.

Clarke & D.T. Lyons (eds.) Recent Themes in the Philosophy of Science:

Scientific Realism and Common Sense. Dordrecht: Kluwer.

Norton, D.F. (ed.) (1993) The Cambridge Companion to Hume. Cambridge:

Cambridge University Press.

Odifreddi, P. (1997) “Review of Shadows of the Mind by R. Penrose”, The

Journal of Symbolic Logic, 62(2), 673-75.

Okasha, S. (1997) “Laudan and Leplin on Empirical Equivalence”, British

Journal for the Philosophy of Science, 48(2), 251-6.

374

O'Leary-Hawthorne, J. (1994) “What Does van Fraassen's Critique of Scientific

Realism Show?” The Monist, 77, 128-45.

Onuf, N. (1989) World of our making: Rules and rule in social theory and

International Relations. Columbia, SC: University of South Carolina Press.

Onuf, N. (2012) “Three Generations of IR”, Unpublished Lecture Notes,

History & Philosophy of IR Seminar, Tinos, Greece.

Outhwaite, W. (ed.) (2003) The Blackwell Dictionary of Modern Social

Thought, 2nd edition. Oxford: Blackwell.

Palmer, R.E. (ed.) (2001) Gadamer in Conversation – Reflections and

Commentary. New Haven, London: Yale University Press.

Papineau, D. (ed.) (1996) The Philosophy of Science. Oxford: Oxford

University Press.

Pappas, G.S. (2000) Berkeley's Thought. Ithaca: Cornell University Press.

Parusnikova, Z. (1992) “Is a Postmodern Philosophy of Science Possible?”,

British Journal for the Philosophy of Science, 23(1), 21-37.

Patomäki, H. (2002) After International Relations – Critical Realism and the

(Re)construction of World Politics. London, New York: Routledge.

Patomäki, H. & Wight, C. (2000) “After Postpositivism: The Promises of

Critical Realism”, International Studies Quarterly, 44(2), 213-37.

Paul, H. (2008) Introduction to Quantum Theory. Cambridge: Cambridge

University Press.

Pickering, A. (1995) The Mangle of Practice. Chicago: University of Chicago

Press.

Penrose, R. (1989) The Emperor’s New Mind - Concerning Computers, Minds

and the Laws of Physics. Oxford: Oxford University Press.

375

Penrose, R. (1994a) Shadows of the Mind - A search for the missing science of

consciousness. Oxford: Oxford University Press.

Penrose, R. (1994b) Interview with Jane Clark, Journal of Consciousness

Studies, 1(1), 17-24.

Penrose, R. (1994c) “Mechanisms, Microtubules and the Mind”, Journal of

Consciousness Studies, 1(2), 241-49.

Penrose, R. (1995) “Beyond the doubting of a Shadow – A reply to

commentaries on Shadows of the Mind”, PSYCHE – An Interdisciplinary

Journal of Research on Consciousness, available online at

http://www.theassc.org/files/assc/2338.pdf.

Penrose, R. (1997) “On Understanding Understanding”, International Studies

in the Philosophy of Science, 11(1), 7-20.

Penrose, R. (2004) The Road to Reality – A Complete Guide to the Laws of the

Universe. London: Jonathan Cape.

Penrose, R. (2011) “Gödel, the Mind and the Laws of Physics”, in M. Baaz et

al. (eds.) Kurt Gödel and the Foundations of Mathematics. Cambridge:

Cambridge University Press.

Penrose, R, Shimony, A, Cartwright, N., & Hawking, S. (1997) The large, the

small and the human mind. Cambridge: Cambridge University Press.

Pinker, S. (2011) The Better Angels of Our Nature – The Decline of Violence in

History and its Causes. London: Allen Lane.

Pitcher, G. (1977) Berkeley. London: Routledge.

Poincaré, H. ([1905]1952) Science and Hypothesis. New York: Dover.

Politis, V. (2004) Routledge Philosophy Guidebook to Aristotle and the

Metaphysics. London, New York: Routledge.

376

Popper, K. ([1957] 1997) The Poverty of Historicism. London: Routledge.

Popper, K. ([1963] 1996) Conjectures and Refutations – The Growth of

Scientific Knowledge. London: Routledge.

Popper, K. (1994) The Myth of the Framework: In Defence of Science and

Rationality (edited by M. A. Notturno). London, New York: Routledge.

Porpora, D. (2007) Book Review [of Wight 2006], Journal of Critical Realism,

6(2), 305-12.

Preston, J. (1998) “Science as Supermarket: ‘Post-Modern’ Themes in Paul

Feyerabend’s Later Philosophy of Science”, Studies in the History and

Philosophy of Science, 29, 425-47.

Pribram, K. (1971) Languages of the Brain. New Jersey: Prentice Hall.

Pribram, K. (1986) “The Cognitive Revolution and the Mind/Brain Issues”,

American Psychologist, 41(5), 507-520.

Psillos, S. (1994) “A Philosophical Study of the Transition from the Caloric

Theory of Heat to thermodynamic: Resisting the Pessimistic Meta-

Induction”, Studies in the History and Philosophy of Science, 25, 159-90.

Psillos, S. (1995) “Is Structural Realism the Best of Both Worlds?”, Dialectica,

49(1), 15-46.

Psillos, S. (1996) “Scientific Realism and the ‘Pessimistic Induction’”,

Philosophy of Science, 63(5), 306-14.

Psillos, S. (1997) “Kitcher on Reference”, International Studies in the

Philosophy of Science, 11(3), 259-72.

Psillos, S. (1999) Scientific Realism: How Science Tracks Truth. London:

Routledge.

377

Psillos, S. (2000a) “The Present State of the Scientific Realism Debate”, British

Journal for the Philosophy of Science, 51, 705-28 • Also in S. Psillos

Knowing the Structure of Nature – Essays on Realism and Explanation. New

York: Palgrave Macmillan.

Psillos, S. (2000b) “Agnostic Empiricism versus Scientific Realism: Belief in

Truth Matters”, International Studies in the Philosophy of Science, 14(1),

57-75.

Psillos, S. (2001a) “Is Structural Realism Possible?”, Philosophy of Science,

68(3), S13-S24 • Also in S. Psillos Knowing the Structure of Nature –

Essays on Realism and Explanation. New York: Palgrave Macmillan.

Psillos, S. (2001b) “Predictive Similarity and the Success of Science: a Reply to

Stanford”, Philosophy of Science, 68(3), 346-55.

Psillos, S. (2005) “Scientific Realism and Metaphysics”, Ratio (new series),

18(4), 385-404 • Also in S. Psillos Knowing the Structure of Nature – Essays

on Realism and Explanation. New York: Palgrave Macmillan.

Psillos, S. (2009) Knowing the Structure of Nature – Essays on Realism and

Explanation. New York: Palgrave Macmillan.

Putnam, H. (1975a) Philosophical Papers, vol.1: Mathematics, Matter and

Method. Cambridge: Cambridge University Press.

Putnam, H. (1975b) Philosophical Papers, vol. 2: Mind, Language and Reality.

Cambridge: Cambridge University Press.

Putnam, H. (1975c) “The Meaning of ‘Meaning’”, in H. Putnam Philosophical

Papers, vol. 2: Mind, Language and Reality. Cambridge: Cambridge

University Press.

378

Putnam, H. (1978) Meaning and the Moral Sciences. London: Routledge and

Kegan Paul.

Putnam, H. (1981) Reason, Truth and History. Cambridge: Cambridge

University Press.

Putnam, H. (1982) “Three Kinds of Scientific Realism”, Philosophical

Quarterly, 32(128), 195-200 • Also in H. Putnam Words & Life (edited by J.

Conant). Cambridge, Massachusetts: Harvard University Press.

Putnam, H. (1983a) Philosophical Papers, vol.3: Realism and Reason.

Cambridge: Cambridge University Press.

Putnam, H (1983b) “Why There Isn't a Ready Made World”, in H. Putnam

Words & Life (edited by J. Conant). Cambridge, Massachusetts: Harvard

University Press.

Putnam, H. (1983c) “Reference and Truth”, in H. Putnam Philosophical

Papers, vol.3: Realism and Reason. Cambridge: Cambridge University

Press.

Putnam, H. (1995) Words & Life (edited by J. Conant). Cambridge,

Massachusetts: Harvard University Press.

Quine, W.O. (1960) Word and Object. Cambridge, Massachusetts: MIT Press.

Quine, W.O. (1975) “On Empirically Equivalent Systems of the World”,

Erkenntnis, 9, 313-28.

R.J.B. (1980) Book Review [of the second edition (by the Harvester Press, in

1978) of Bhaskar [1975] 2008], Review of Metaphysics, 33(3), 619-620.

Radder, H. (1993) “Science, Realization and Reality: The Fundamental Issues”,

Studies in the History and Philosophy of Science, 24(3), 327-49.

379

Rawls, J. (2001) The Law of Peoples, new edition. Cambridge, Massachusetts:

Harvard University Press.

Redhead, M. (1995) From Physics to Metaphysics. Cambridge: Cambridge

University Press.

Redhead, M. (2001) “Review Symposium: The Quests of a Realist”,

Metascience, 10(3), 341-47.

Redhead, M. (2007) “The Unseen World”, in C. Cheyne & J. Worrall (eds.)

Rationality and Reality: Conversations with Alan Musgrave. Dordrecht:

Springer.

Rehg, W. (2009) Cogent Science in Context: The Science Wars, Argumentation

Theory, and Habermas. Cambridge: MIT Press.

Rengger, N. (2011) “The world turned upside down? – Human rights and

International Relations after 25 years”, International Affairs, 87(5), 1159-

1178.

Rescher, N. (ed.) (1987) Scientific Inquiry in Philosophical Perspective.

Lanham: University Press of America.

Resnik, D.B. (1994) “Hacking’s Experimental Realism”, Canadian Journal of

Philosophy, 24(3), 395-412, • Also in M. Curd & J.A. Cover (eds.) The

Philosophy of Science: The Central Issues. London, New York: W.W.

Norton & Co.

Resnik, M.D. (1995) “Scientific vs. Mathematical Realism: The

Indispensability Argument”, Philosophia Mathematica, III(3), 166-74.

Reus-Smit, C. (2012) “International Relations, Irrelevant? Don’t Blame

Theory”, Millennium: Journal of International Studies, 40(3), 525-40.

380

Rich, M. (1999) “Essay Review [of Leplin 1997a]: Realism’s New Miracle?”,

Studies in the History and Philosophy of Science, 30(3), 511-21.

Risjord, M. (1998), “Norms and Explanation in the Social Sciences”, Studies in

the History and Philosophy of Science, 29(2), 223-37.

Ringmar, E. ([1997] 2001) “Alexander Wendt: a social scientist struggling with

history”, in I.B. Neumann & O. Wæver (eds.) The Future of International

Relations - Masters in the Making. London, New York: Routledge.

Rosen, G. (1994) “What is Constructive Empiricism?”, Philosophical Studies,

74(2), 143-78.

Rosenberg, A. (1996) “A Field Guide to Recent Species of Naturalism”, British

Journal for the Philosophy of Science, 47(1), 1-29.

Roth, P.A. (1996) “Will the Real Scientists Please Stand Up? Dead Ends and

Live Issues in the Explanation of Scientific Knowledge”, Studies in the

History and Philosophy of Science, 27(1), 43-68.

Roth, P.A. (2003) “Kitcher’s Two Cultures”, Philosophy of the Social Sciences,

33(3), 386-405.

Rouse, J. (1981) “Kuhn, Heidegger, and Scientific Realism”, Man and World,

14(3), 269-90.

Rouse, J. (1987) Knowledge and Power: Toward a Political Philosophy of

Science. Ithaca: Cornell University Press.

Rouse, J. (1991a) “Philosophy of Science and the Persistent Narratives of

Modernity”, Studies in the History and Philosophy of Science, 22(1), 141-62.

Rouse, J. (1991b) “The Politics of Postmodern Philosophy of Science”,

Philosophy of Science, 58(4), 607-27.

381

Ruetsche, L. (2002a) “Interpreting Quantum Field Theory”, Philosophy of

Science, 69(2), 348-78.

Ruetsche, L. (2002b) “Interpreting Quantum Theories”, in P. Machamer & M.

Silberstein (eds.) The Blackwell Guide to the Philosophy of Science. Malden,

MA: Blackwell.

Ruetsche, L. (2011) Interpreting Quantum Mechanics. Oxford: Oxford

University Press.

Ruse, M. (1981) Book Review [of Bhaskar 1979], ISIS, 72(3), 493-95.

Ryner, J.M. (2002) Capitalist Restructuring: Globalisation and the Third Way.

London: Routledge.

Saatsi, J. (2005) “Reconsidering the Fresnel-Maxwell Shift: How the Realist

Can Have Her Cake and EAT It Too”, Studies in the History and Philosophy

of Science, 36(3), 509-38.

Salmon, M.H. (2007) Introduction to Logic and Critical Thinking, 5th edition.

Belmont, CA: Thomson Wadsworth.

Salmon, M.H. et al. ([1992]1999) Introduction to the Philosophy of Science.

Indianapolis, Cambridge: Hackett Publishing Company.

Salmon, W.C. ([1992]1999) “Scientific Explanation”, in M.H. Salmon, et al.

Introduction to the Philosophy of Science. Indianapolis, Cambridge: Hackett

Publishing Company.

Savage, W. (ed.) (1990) Minnesota Studies in the Philosophy of Science, vol.

14: Scientific Theories. Minneapolis: University of Minnesota Press.

Sayer, A. (2000) Realism and Social Science. London: Sage Publications.

Schaffner, K.F. (1992) Introduction to the Philosophy of Science. Englewood

Cliffs, NJ: Prentice Hall.

382

Schattschneider, D. (1990), Visions of Symmetry: Notebooks, Periodic

Drawings, and Related Work by M. C. Escher. New York: Freeman.

Schatzki, T.R. (1991) “Elements of a Wittgensteinian Philosophy of the

Human Sciences”, Synthese, 87(2), 311-29.

Schmidt, B.C. (1997) “Further Ahead or Further Behind? The Debate over

Positivism” – Book Review [of Nicholson 1996a & Neufeld 1995],

Mershon International Studies Review, 41, 107-112.

Searle, J.R. (1995) “The Mystery of Consciousness” – Book Review [of Crick

1995 & Penrose 1994a], The New York Review of Books, XLII (17), 60-66.

Sellars, W. (1963) Science, Perception and Reality. Atascadero, CA:

Ridgeview P.C.

Senechal, M. (1992) Book Review [of Penrose 1989 & Schattschneider 1990],

The Mathematical Intelligencer, 14, 72-77.

Shapiro, I. & Wendt, A. (1992) “The difference that realism makes: social

science and the politics of consent”, Politics and Society, 20(2), 197-223.

Shapiro, S. (1997) Philosophy of Mathematics - Structure and Ontology.

Oxford: Oxford University Press.

Shapiro, S. (1998) “Incompleteness, Mechanism, and Optimism”, The Bulletin

of Symbolic Logic, 4(3), 273-302.

Shermer, S. (2004) “Quantum Quackery”, Scientific American, 291(6), 38.

Singer, J.D. (1961) “The Level-of-Analysis Problem in International

Relations”, World Politics, 14(1), 77-92.

Smart, J.J.C. (1963) Philosophy and Scientific Realism. London: Routledge &

Kegan Paul.

Sibeon, R. (2004) Rethinking Social Theory. London: Sage Publications.

383

Smith, P. (1998) “Approximate Truth and Dynamical Theories”, British

Journal for the Philosophy of Science, 49(2), 253-77.

Smith, S. (1995) “The Self-Images of a Discipline: A Genealogy of

International Relations Theory”, in K. Booth & S. Smith, S. (eds.)

International Relations Theory Today. Cambridge: Polity Press.

Smith, S. (1996) “Positivism and Beyond”, in S. Smith, K. Booth & M. Zalewski

(eds.) International Theory: Positivism and Beyond. Cambridge: Cambridge

University Press.

Smith, S. (1997) “Power and truth: a reply to William Wallace”, Review of

International Studies, 23(4), 507-16.

Smith, S. (2000) “Wendt’s World”, Review of International Studies, 26(1), 151-

63.

Smith, S. (2007) “Introduction: Diversity and Disciplinarity in International

Relations Theory”, in T. Dunne, M. Kurki & S. Smith (eds.) International

Relations Theory – Discipline and Diversity, 2nd edition. Oxford: Oxford

University Press.

Smith, S., Booth, K. & Zalewski M. (eds.) (1996) International Theory:

Positivism and Beyond. Cambridge: Cambridge University Press.

Smith, S., Hadfield, A. & Dunne, T. (eds.) (2008) Foreign Policy - Theories,

Actors, Cases, Oxford: Oxford University Press.

Snidal, D. & Wendt, A. (2009) “Why there is International Theory now”,

International Theory, 1(1), 1-14.

Stanford, P.K. (2000) “An Antirealist Explanation of the Success of Science”,

Philosophy of Science, 67(2), 266-84.

384

Stanford, P.K. (2001) “Refusing the Devil’s Bargain: What Kind of

Underdetermination Should We Take Seriously?”, Philosophy of Science,

68(3), S1-S12.

Stanford, P.K. (2003a) “Pyrrhic Victories for Scientific Realism”, Journal of

Philosophy, 100(11), 553-72.

Stanford, P.K. (2003b) “No Refuge for Realism: Selective Confirmation and

the History of Science”, Philosophy of Science, 70(5), 913-25.

Stanford, P.K. (2006) “Darwin’s Pangenesis and the Problem of Unconceived

Alternatives”, British Journal for the Philosophy of Science, 57(1), 121-44.

Stapp, H.P. (1999) “Attention, Intention, and Will in Quantum Physics”,

Journal of Consciousness Studies, 6(8-9), 143-164.

Stein, H. (1989) “Yes, but… Some Sceptical Remarks on Realism and Anti-

Realism”, Dialectica, 43(1-2), 47-65.

Steiner, M. (1983) “The search for order in a disorderly world: worldviews and

prescriptive decision paradigms”, International Organization, 37(3), 373-

413.

Stenger, V.J. (1996) The Unconscious Quantum – Metaphysics in Modern

Physics and Cosmology. New York: Prometheus Book.

Stump, E. (2003) Aquinas. London, New York: Routledge.

Suganami, H. (2001) “Alexander Wendt and the English School”, Journal of

International Relations and Development, 4(4), 403-23.

Suganami, H. (2002) “On Wendt’s Philosophy: A Critique”, Review of

International Studies, 28 (1), 23-37.

385

Suganami, H. (2006) “Wendt, IR, and Philosophy: A Critique”, in S. Guzzini &

A. Leander (eds.) Constructivism and International Relations – Alexander

Wendt and his critics. London, New York: Routledge.

Susskind, L. (2008) The Black Hole War - My battle with Stephen Hawking to

Make the World Safe. London: Little Brown and Company.

Susskind, L. (2011) “Bad Boy of Physics” (Interview by Peter Byrne),

Scientific American, 305(1), 64-67.

Talbot, M. (2011) The Holographic Universe: The Revolutionary Theory of

Reality. New York: Harper Perennial.

Teller, P. (2001) “Whither Constructive Empiricism?”, Philosophical Studies,

106(1-2), 123-50.

Tieszen, R. (1996) “Review of Shadows of the Mind, by Roger Penrose”,

Philosophia Mathematica, 4(3), 281-290.

Tieszen, R. (2005) “Consciousness of abstract objects”, in D. Smith and A.

Thomason (eds.), Phenomenology and the Philosophy of Mind. Oxford:

Oxford University Press.

Tieszen, R. (2011) After Gödel – Platonism and Rationalism in Mathematics

and Logic. Oxford: Oxford University Press.

Tonra, B. (2001) The Europeanization of National Foreign Policy – Dutch,

Danish and Irish Foreign Policy in the European Union. Hampshire:

Ashgate.

Tonra, B & Christiansen, T. (2004) (eds.) Rethinking European Foreign

Policy. Manchester: Manchester University Press.

Trout, J.D. (1994) “A Realistic Look Backward”, Studies in the History and

Philosophy of Science, 25(1), 37-64.

386

Tuck, R. ([1999] 2001) The Rights of War and Peace – Political Thought and

the International Order from Grotius to Kant. Oxford: Oxford University

Press.

Turner, D.D. (2004) “The Past vs. the Tiny: Historical Science and the

Abductive Arguments for Realism”, Studies in the History and Philosophy

of Science, 35(1), 1-17.

Van Fraassen, B.C. (1976) “To Save the Phenomena”, Journal of Philosophy,

73(18), 623-32.

Van Fraassen, B.C. (1980) The Scientific Image. Oxford: Clarendon Press.

Van Fraassen, B.C. ([1985] 2000) “Empiricism in the Philosophy of Science”,

in P.M. Churchland & C.A. Hooker (eds.) Images of Science: Essays on

Realism and Empiricism, with a Reply from Bas C. van Fraassen. Chicago:

University of Chicago Press.

Van Fraassen, B.C. (1989) Laws and Symmetry. Oxford: Clarendon Press.

Van Fraassen, B.C. (1994) “Gideon Rosen on Constructive Empiricism”,

Philosophical Studies, 74(2), 179-92.

Van Loocke, P. (ed.) (2001) The Physical Basis of Consciousness. Philadelphia:

John Benjamins.

Vincent, R.J. (1974) Nonintervention and International Order. Princeton, New

Jersey: Princeton University Press.

Vincent, R.J. (1986) Human Rights and International Relations. Cambridge:

Cambridge University Press for the Royal Institute of International Affairs.

Vitiello, G. (1995) “Dissipation and Memory Capacity in the Quantum Brain

Model”, International Journal of Modern Physics B, 9, 973-989.

Vitiello, G. (2001) My Double Unveiled. Amsterdam: Benjamins.

387

Vitiello, G. (2002) “Dissipative Quantum Brain Dynamics”, in K. Yasue et al.

(eds.) No Matter, Never Mind. Amsterdam: Benjamins.

Wagner, B. & Gebauer, J. (2008) Alexander Wendt’s auto-critique and social

construction, Draft version: 21, available online at

http://eui.academia.edu/BenWagner/Papers/185784/Alexander_Wendt_s_au

to-critique_and_social_constructivism.

Walker, R.B.J. (1991) “State Sovereignty and the Articulation of Potential

Space/time”, Millennium: Journal of International Studies, 20(3), 445-461.

Walker, R.B.J. (1993) Inside/Outside: International Relations as Political

Theory. Cambridge: Cambridge University Press.

Wallace, W. (1996) “Truth powers, monks and technocrats: theory and practice

in international relations”, Review of International Studies, 22(3), 301-21.

Wallerstein, I. (1974) The Modern World-System, vol. I. New York: Academic

Press.

Wallerstein, I. (2004) World Systems Analysis: An Introduction. Durham,

London: Duke University Press.

Walsh, W.H. ([1977] 2008) Philosophy of History - An Introduction. Delhi:

Cosmo Publications.

Waltz, K.N. (1979) Theory of International Politics. New York: McGraw-Hill,

Inc.

Wang, H. (1974) From Mathematics to Philosophy. New York: Humanities

Press.

Weber, M. (2012) “Ontologies, depth, and otherwise: critical notes on Wight’s

meta-theoretical proposal of a scientific realist IR”, Review of international

Studies, 38(1), 223-34.

388

Weinberg, S. (1994) Dreams of a Final Theory. New York: Vintage Books.

Weingard, R. (1988) "A philosopher looks at string theory" in PSA: Proceedings

of the Biennial Meeting of the Biennial Meeting of the 1988 Philosophy of

Science Association, Vol. 2: Symposia and Invited Papers. East Lansing, MI:

Philosophy of Science Association • Also in C. Callender & N. Huggett (eds.)

Physics Meets Philosophy at the Planck Scale: Contemporary Theories in

Quantum Gravity. Cambridge: Cambridge University Press.

Wendt, A. (1987) “The Agent-Structure Problem in International Relations

Theory”, International Organization, 41(3), 335-70.

Wendt, A. (1991) “Bridging the theory/meta-theory gap in international

relations”, Review of International Studies, 17 (4), 383-392.

Wendt, A. (1992) “Anarchy is what states make of it: The social construction of

power politics”, International Organization, 46(2), 391-425.

Wendt, A. (1992a) “Levels of analysis vs. agents and structures: part III”,

Review of International Studies, 18(2), 181-85.

Wendt, A. (1995) “Constructing International Politics”, International

Organization, 20(1), 71-81.

Wendt, A. (1998) “On Constitution and Causation in International Relations”,

Review of International Studies, 24(5), 101-17.

Wendt, A. (1999) Social Theory of International Politics. Cambridge:

Cambridge University Press.

Wendt, A. (2000) “On the Via Media: a response to the critics”, Review of

International Studies, 26(1), 165-180.

Wendt, A. (2003) “Why a World State is Inevitable”, European Journal of

International Relations, 9(4), 491-542.

389

Wendt, A. (2004) “The state as a person in international theory”, Review of

International Studies, 30(2), 289-316.

Wendt, A. (2006a) “Social Theory as a Cartesian Science: An Auto-Critique

from a Quantum Perspective”, in Guzzini & Leander (eds.) Constructivism

and International Relations – Alexander Wendt and his critics. London, New

York: Routledge.

Wendt, A. (2006b) “Quantum Mind and Social Science”, available at

http://www.quantumconsciousness.org.views/QuantumMindAndSocialScie

nce.html.

Wendt, A. (2010) “Flatland: Quantum Mind and the International Hologram”,

in M. Albert et al. (eds.) New Systems Theories of World Politics. New

York: Palgrave Macmillan.

Wessels, L. (1993) “Scientific Realism and Quantum Mechanical Realism”,

Midwest Studies in Philosophy, 18, 317-31.

Weston, T. (1992) “Approximate Truth and Scientific Realism”, Philosophy of

Science, 59(1), 53-74.

Wheeler, J.A. & Zurek, W.H. (eds.) (1983) Quantum Theory and Measurement.

Princeton. New Jersey: Princeton University Press.

Whitbeck, C. (1977) Book Review [of Bhaskar ([1975] 2008)], Philosophical

Review, 86(1), 114-18.

White, A.R. (ed.) ([1968] 1973) The Philosophy of Action. Oxford: Oxford

University Press.

White, B. (2001) Understanding European Foreign Policy. New York:

Palgrave.

390

White, B. (2004) “Foreign policy analysis and European foreign policy”, in B.

Tonra & T. Christiansen (eds.) Rethinking European Foreign Policy.

Manchester: Manchester University Press.

Whitham, B. (2012) “Critical Realism and International Relations: Causal

Explanations for Liberal War”, available online at:

http://www.academia.edu/348746/Critical_Realism_and_International_Rela

tions_Causal_Explanations_for_Liberal_War (Paper presented at the 2012

International BISA-ISA Conference in Edinburgh, Friday 22nd June 2012).

Wight, C. (1998) Review Essay: “Philosophical Geographies: Navigating

Philosophy in Social Science”, Philosophy of the Social Sciences, 28(4),

552-66.

Wight, C. (1999) “They Shoot Dead Horses Don’t They? Locating Agency in the

Agent-Structure Problematique”, European Journal of International Relations,

5(1), 109-42.

Wight, C. (2002) “Philosophy of Social Science and International Relations”, in

W. Carlsnaes, T. Risse, & B.A. Simmons (eds.) Handbook of International

Relations. London: Sage Publications.

Wight, C. (2003) Book Review [of Patomäki 2002], Cooperation and Conflict,

38(3), 323-24.

Wight, C. (2004) “State agency: social action without human activity?”, Review of

International Studies, 30(2), 269-80.

Wight, C. (2006) Agents, Structures and International Relations: Politics as

Ontology. Cambridge: Cambridge University Press.

391

Wight, C. (2007) “A Manifesto for Scientific Realism in IR: Assuming the

Can-Opener Won’t Work”, Millennium: Journal of International Studies,

35(2), 379-98.

Wight, C. (2007a) “Inside the epistemological cave all bets are off”, Journal of

International Relations and Development, 10(1), 40-56.

Wight, C. (2007b) “A response to Friedrich Kratochwil: why shooting the

messenger does not make the bad news go away!”, Journal of International

Relations and Development, 10(3), 301-15.

Wight, C. (2008) “Living with Hume’s Problem: Reply to Patrick Jackson”,

Cooperation and Conflict, 43(3), 357-60.

Wight, C. (2012) “Critical realism: some responses”, Review of International

Studies, 38(1), 267-74.

Wight, C. & Joseph, J. (2010) “Scientific Realism and International Relations”,

in J. Joseph & C. Wight (eds.) Scientific Realism and International

Relations. New York: Palgrave Macmillan.

Wilson, M. (1980) “The Observational Uniqueness of Some Theories”, Journal

of Philosophy, 77(4), 208-33.

Winch, P. (1958) The Idea of a Social Science and its Relation to Philosophy,

2nd edition. London: Routledge.

Winkler, K.P. (1989) Berkeley: An Interpretation. Oxford: Clarendon Press.

Wœver, O. (1994) “Resisting the Temptation of Post Foreign Policy Analysis”,

in W. Carlsnaes & S. Smith (eds.) European Foreign Policy – The EC and

Changing Perspectives in Europe. London: Sage Publications.

Wong, R. and Hill, C. (eds.) (2011a) National and European Foreign Policies

– Towards Europeanization. London: Routledge.

392

Wong, R. and Hill, C. (2011b) “Introduction”, in R. Wong & C. Hill (eds.)

National and European Foreign Policies – Towards Europeanization.

London: Routledge.

Worrall, J. (1989) “Structural Realism: The Best of Both Worlds?”, Dialectica,

43(1-2), 99-124 • Also in D. Papineau (ed.) The Philosophy of Science.

Oxford: Oxford University Press, and in J. Worrall (ed.) The Ontology of

Science. Aldershot: Dartmouth.

Worrall, J. (ed.) (1994a) The Ontology of Science. Aldershot: Dartmouth.

Worrall, J. (1994b) “How to Remain (Reasonably) Optimistic: Scientific

Realism and the ‘Luminiferous Ether’”, in D. Hull, M. Forbes & R.M.

Burian (eds.) PSA 1994, vol. 1. East Lansing, MI: Philosophy of Science

Association.

Worrall, J. (1999) “Two Cheers for Naturalised Philosophy of Science – or: Why

Naturalised Philosophy of Science is Not the Cat’s Whiskers”, Science &

Education, 8(4), 339-61.

Worrall, J. (2007). “Miracles and Models: Why Reports of the Death of

Structural Realism may be Exaggerated”, Royal Institute of Philosophy

Supplements, 61, 125-54.

Wylie, A. (1986) “Arguments for Scientific Realism: the Ascending Spiral”,

American Philosophical Quarterly, 23(3), 287-97.

Yalvaç, F. (2010) “Critical Realism, International Relations Theory and

Marxism”, in J. Joseph & C. Wight (eds.) Scientific Realism and

International Relations. New York: Palgrave Macmillan.

Yasue, K., et al. (eds.) (2002.), No Matter, Never Mind. Amsterdam:

Benjamins.

393

Zahar, E. (2001), Poincaré’s Philosophy: From Conventionalism to

Phenomenology. Chicago: Open Court.

Zehfuss, M. (2002) Constructivism in International Relations: The Politics of

Reality. Cambridge: Cambridge University Press.

Zukay, G. (1979) The Dancing Li Masters. New York: Quill.